library(dplyr)
library(tidyverse)
library(ggplot2)
library(readr)
library(stringr)
library(naniar)
library(foreign)
library(plm)
library(tigris)
library(leaflet)
library(sf)
library(tmap)Final Project for Advanced Methods
In this study we seek to find connections between two population processes: mortality and fertility. Deaths of despair, first introduced by Case & Deaton (2017) as deaths from suicide, overdoses, and alcohol related injuries, have been on the rise since the 1950s. This phenomenon is seen within non-Hispanic white populations with low levels of education. (Case & Deaton 2017). Fertility, on the other hand, has been decreasing since 2008. Some research, however, has found increases in births to certain segments of the population. For example, Caudillo & Villareal (2021) find a positive association between opioid overdose deaths, a component of deaths of despair, and nonmarital fertility. This study seeks to expand the scope of Caudillo & Villareal (2021) by focusing on various classifications of state-level mortality and nonmarital fertility.
Theoretical Framework
We operationalize despair from the lens of mortality. We test several different measures of mortality may capture different dimensions of state-level despair; these include death rates for deaths classified as deaths of despair (we use the Case and Deaton ICD code list), opioid death rates and all cause-mortality amongst working age adults.
To understand the relationship between state-level despair and nonmarital fertility, we employ three theoretical frameworks. The first framework is the idea that women postpone childbearing in times of economic uncertainty (Morgan & Bachrach & Morgan 2013; Lutz et al. 2006). Although this argument is economic, we believe it applies to mortality because the deaths we are studying are either at early ages or due to unnatural causes. Witnessing an influx of these deaths may induce feelings of “despair.” The second theory explains nonmarital fertility’s connection to despair by arguing that despair leads to hopelessness. Using this framework, women who feel hopeless seek to have children now rather than postpone childbearing they because they believe their situation is not going to improve anyways (Kearny & Levine 2011).
Lastly, this study uses the concepts of exposure and cumulative disadvantage theory to explain the time-relationship of despair with fertility. Exposure considers which factors a person is exposed to over time to see if it can influence their health outcomes (Merton 1968; Montez & Hayward 2014). Cumulative disadvantage theory explains how an individual’s negative experiences can pile and build off each other to contribute to their poor health (Melchoir et al. 2006). We use the concepts of exposure and cumulative disadvantage in explaining how economic despair can affect nonmarital fertility overtime. For example, a woman may see deaths of despair affecting her community over a certain time period. The longer this period is, the more pronounced the effect of despair on nonmarital fertility decision-making.
We seek to investigate the validity of these theories by testing the association between nonmarital fertility, deaths of despair, opioid deaths, and working age mortality. Our hypotheses are:
H1: Higher rates of deaths classified as deaths of despair will be associated with higher rates of nonmarital fertility
H1a: Length of exposure to deaths of despair will be associated with higher rates of nonmarital fertility
H2: Higher rates of all-cause working age adult mortality will be associated with higher rates of nonmarital fertility
H2a: Length of exposure to all-cause working age mortality will be associated with higher rates of nonmarital fertility
H3: Higher rates of opioid death will be associated with higher rates of nonmarital fertility
H3a: Length of exposure to opioid deaths will be associated with higher rates of nonmarital fertility
Literature Survey
Research has identified a downward trend in nonmarital fertility since 2008. For example, Schneider & Gemmill (2016) use time series analysis and fixed effects methods to study nonmarital fertility using vital statistics and American Community Survey data. They argue that the increased utilization of long-term contraception methods and effects of the Great Recession have led to a decrease in unplanned births. Another project by Kearney & Levine (2015) specifically study trends in the teen birth rate using vital statistic and census data as well. This study also argues that the increase in unemployment resulting from the Great Recession also led to a decrease in teen births. Both studies investigate state-level trends.
Studying nonmarital fertility at smaller geographic levels can provide even more finite detail. As mentioned earlier, Caudillo and Villareal (2021) find support for a positive association between opioid death rates and nonmarital fertility at the CPUMA level. Since nonmarital fertility has been decreasing since 2008, Caudillo & Villareal (2021) argue that if there weren’t such an increase in opioid deaths, nonmarital fertility would have decreased further. To investigate this further we sought out to study the relationship between different classifications of mortality: deaths of despair, opioid deaths and working age mortality at the state-level for this project. Our study adds to the literature by investigating opioid death rates and nonmarital fertility at the state-level as well as other classifications of mortality. Moving forward, we plan to carry out this analysis at the county-level to get into even more detail.
Methods
First, we source mortality rates for deaths of despair, opioid overdose deaths and all-cause working age mortality from CDC wonder for the years 2006-2020 at the state-level. Secondly, we use ACS data from 2010-2021 to calculate state-level nonmarital population estimates. Lastly, we use CDC wonder natality data to get total nonmarried births at the state-level. We then calculate the state-level nonmarital fertility rates. We use the total nonmarital fertility rate as our dependent variable, total nonmarried births over the total nonmarried female population in their childbearing years (15-49). These birth rates are interpreted as births per 1,000 women.
Mortality data is from the CDC WONDER compressed mortality and multiple cause of death files. Midlife Mortality is defined as a death from any cause for an individual between the ages of 25 and 64 (this is the most common age range see Woolf 2018, Woolf 2019, Montez 2022). For deaths of despair we use the ICD classification scheme introduced by Case and Deaton for their cause of death definitions. Opioid death classifications follow the from the work of Caudillo and Villareal. All mortality rates are age-standardized.
Our unit of analysis is state-year. We set the lag between the independent variables, mortality rates and dependent variable, nonmarital fertility rates from 1-5 years to see the differences in the effect of exposure. We use the Hausman test to determine if random effects or fixed effects is the superior specification for each lag. Fixed effects controls for unobserved factors that may be related to either state-level mortality or fertility, while a random effects model allows these unobserved factors to vary.
1. Load Libraries
2. Import DOD Data
DOD05 <- read_delim("DOD05.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD06 <- read_delim("DOD06.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD07 <- read_delim("DOD07.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD08 <- read_delim("DOD08.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD09 <- read_delim("DOD09.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD10 <- read_delim("DOD10.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD11 <- read_delim("DOD11.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD12 <- read_delim("DOD12.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD13 <- read_delim("DOD13.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD14 <- read_delim("DOD14.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD15 <- read_delim("DOD15.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
DOD16 <- read_delim("DOD16.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
#example of data frame layout
head(DOD16)# A tibble: 6 × 7
Notes State `State Code` Deaths Population `Crude Rate` `Age Adjusted Rate`
<chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
1 <NA> Alabama 01 1902 4863300 39.1 38.1
2 <NA> Alaska 02 490 741894 66 64.1
3 <NA> Arizona 04 3862 6931071 55.7 53.8
4 <NA> Arkansas 05 1188 2988248 39.8 38.9
5 <NA> Califor… 06 13683 39250017 34.9 32.7
6 <NA> Colorado 08 2901 5540545 52.4 50 3. Clean DOD Data
#05
DOD05$Notes <- NULL
DOD05 <- DOD05 %>% drop_na()
DOD05$Y05 <- DOD05$`Age Adjusted Rate`
DOD05$Deaths <- NULL
DOD05$Population <- NULL
DOD05$`Crude Rate` <- NULL
DOD05$`Age Adjusted Rate` <- NULL
#06
DOD06$Notes <- NULL
DOD06 <- DOD06 %>% drop_na()
DOD06$Y06 <- DOD06$`Age Adjusted Rate`
DOD06$Deaths <- NULL
DOD06$Population <- NULL
DOD06$`Crude Rate` <- NULL
DOD06$`Age Adjusted Rate` <- NULL
#07
DOD07$Notes <- NULL
DOD07 <- DOD07 %>% drop_na()
DOD07$Y07 <- DOD07$`Age Adjusted Rate`
DOD07$Deaths <- NULL
DOD07$Population <- NULL
DOD07$`Crude Rate` <- NULL
DOD07$`Age Adjusted Rate` <- NULL
#08
DOD08$Notes <- NULL
DOD08 <- DOD08 %>% drop_na()
DOD08$Y08 <- DOD08$`Age Adjusted Rate`
DOD08$Deaths <- NULL
DOD08$Population <- NULL
DOD08$`Crude Rate` <- NULL
DOD08$`Age Adjusted Rate` <- NULL
#09
DOD09$Notes <- NULL
DOD09 <- DOD09 %>% drop_na()
DOD09$Y09 <- DOD09$`Age Adjusted Rate`
DOD09$Deaths <- NULL
DOD09$Population <- NULL
DOD09$`Crude Rate` <- NULL
DOD09$`Age Adjusted Rate` <- NULL
#10
DOD10$Notes <- NULL
DOD10 <- DOD10 %>% drop_na()
DOD10$Y10 <- DOD10$`Age Adjusted Rate`
DOD10$Deaths <- NULL
DOD10$Population <- NULL
DOD10$`Crude Rate` <- NULL
DOD10$`Age Adjusted Rate` <- NULL
#11
DOD11$Notes <- NULL
DOD11 <- DOD11 %>% drop_na()
DOD11$Y11 <- DOD11$`Age Adjusted Rate`
DOD11$Deaths <- NULL
DOD11$Population <- NULL
DOD11$`Crude Rate` <- NULL
DOD11$`Age Adjusted Rate` <- NULL
#12
DOD12$Notes <- NULL
DOD12 <- DOD12 %>% drop_na()
DOD12$Y12 <- DOD12$`Age Adjusted Rate`
DOD12$Deaths <- NULL
DOD12$Population <- NULL
DOD12$`Crude Rate` <- NULL
DOD12$`Age Adjusted Rate` <- NULL
#13
DOD13$Notes <- NULL
DOD13 <- DOD13 %>% drop_na()
DOD13$Y13 <- DOD13$`Age Adjusted Rate`
DOD13$Deaths <- NULL
DOD13$Population <- NULL
DOD13$`Crude Rate` <- NULL
DOD13$`Age Adjusted Rate` <- NULL
#14
DOD14$Notes <- NULL
DOD14 <- DOD14 %>% drop_na()
DOD14$Y14 <- DOD14$`Age Adjusted Rate`
DOD14$Deaths <- NULL
DOD14$Population <- NULL
DOD14$`Crude Rate` <- NULL
DOD14$`Age Adjusted Rate` <- NULL
#15
DOD15$Notes <- NULL
DOD15 <- DOD15 %>% drop_na()
DOD15$Y15 <- DOD15$`Age Adjusted Rate`
DOD15$Deaths <- NULL
DOD15$Population <- NULL
DOD15$`Crude Rate` <- NULL
DOD15$`Age Adjusted Rate` <- NULL
#16
DOD16$Notes <- NULL
DOD16 <- DOD16 %>% drop_na()
DOD16$Y16 <- DOD16$`Age Adjusted Rate`
DOD16$Deaths <- NULL
DOD16$Population <- NULL
DOD16$`Crude Rate` <- NULL
DOD16$`Age Adjusted Rate` <- NULL
#example of cleaned data frame
head(DOD16)# A tibble: 6 × 3
State `State Code` Y16
<chr> <chr> <dbl>
1 Alabama 01 38.1
2 Alaska 02 64.1
3 Arizona 04 53.8
4 Arkansas 05 38.9
5 California 06 32.7
6 Colorado 08 50 4. Import Midlife Mort Data
MLM05 <- read_delim("MLM05.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM06 <- read_delim("MLM06.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM07 <- read_delim("MLM07.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM08 <- read_delim("MLM08.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM09 <- read_delim("MLM09.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM10 <- read_delim("MLM10.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM11 <- read_delim("MLM11.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM12 <- read_delim("MLM12.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM13 <- read_delim("MLM13.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM14 <- read_delim("MLM14.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM15 <- read_delim("MLM15.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
MLM16 <- read_delim("MLM16.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
#example of data frame layout
head(MLM16)# A tibble: 6 × 7
Notes State `State Code` Deaths Population `Crude Rate` `Age Adjusted Rate`
<chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
1 <NA> Alabama 01 14652 2522310 581. 502.
2 <NA> Alaska 02 1671 402087 416. 379.
3 <NA> Arizona 04 13885 3456387 402. 355.
4 <NA> Arkansas 05 8504 1514723 561. 485.
5 <NA> Califor… 06 62779 20961732 300. 269.
6 <NA> Colorado 08 9695 3001518 323 292.5. Clean Midlife Mort Data
#05
MLM05$Notes <- NULL
MLM05 <- MLM05 %>% drop_na()
MLM05$Y05 <- MLM05$`Age Adjusted Rate`
MLM05$Deaths <- NULL
MLM05$Population <- NULL
MLM05$`Crude Rate` <- NULL
MLM05$`Age Adjusted Rate` <- NULL
#06
MLM06$Notes <- NULL
MLM06 <- MLM06 %>% drop_na()
MLM06$Y06 <- MLM06$`Age Adjusted Rate`
MLM06$Deaths <- NULL
MLM06$Population <- NULL
MLM06$`Crude Rate` <- NULL
MLM06$`Age Adjusted Rate` <- NULL
#07
MLM07$Notes <- NULL
MLM07 <- MLM07 %>% drop_na()
MLM07$Y07 <- MLM07$`Age Adjusted Rate`
MLM07$Deaths <- NULL
MLM07$Population <- NULL
MLM07$`Crude Rate` <- NULL
MLM07$`Age Adjusted Rate` <- NULL
#08
MLM08$Notes <- NULL
MLM08 <- MLM08 %>% drop_na()
MLM08$Y08 <- MLM08$`Age Adjusted Rate`
MLM08$Deaths <- NULL
MLM08$Population <- NULL
MLM08$`Crude Rate` <- NULL
MLM08$`Age Adjusted Rate` <- NULL
#09
MLM09$Notes <- NULL
MLM09 <- MLM09 %>% drop_na()
MLM09$Y09 <- MLM09$`Age Adjusted Rate`
MLM09$Deaths <- NULL
MLM09$Population <- NULL
MLM09$`Crude Rate` <- NULL
MLM09$`Age Adjusted Rate` <- NULL
#10
MLM10$Notes <- NULL
MLM10 <- MLM10 %>% drop_na()
MLM10$Y10 <- MLM10$`Age Adjusted Rate`
MLM10$Deaths <- NULL
MLM10$Population <- NULL
MLM10$`Crude Rate` <- NULL
MLM10$`Age Adjusted Rate` <- NULL
#11
MLM11$Notes <- NULL
MLM11 <- MLM11 %>% drop_na()
MLM11$Y11 <- MLM11$`Age Adjusted Rate`
MLM11$Deaths <- NULL
MLM11$Population <- NULL
MLM11$`Crude Rate` <- NULL
MLM11$`Age Adjusted Rate` <- NULL
#12
MLM12$Notes <- NULL
MLM12 <- MLM12 %>% drop_na()
MLM12$Y12 <- MLM12$`Age Adjusted Rate`
MLM12$Deaths <- NULL
MLM12$Population <- NULL
MLM12$`Crude Rate` <- NULL
MLM12$`Age Adjusted Rate` <- NULL
#13
MLM13$Notes <- NULL
MLM13 <- MLM13 %>% drop_na()
MLM13$Y13 <- MLM13$`Age Adjusted Rate`
MLM13$Deaths <- NULL
MLM13$Population <- NULL
MLM13$`Crude Rate` <- NULL
MLM13$`Age Adjusted Rate` <- NULL
#14
MLM14$Notes <- NULL
MLM14 <- MLM14 %>% drop_na()
MLM14$Y14 <- MLM14$`Age Adjusted Rate`
MLM14$Deaths <- NULL
MLM14$Population <- NULL
MLM14$`Crude Rate` <- NULL
MLM14$`Age Adjusted Rate` <- NULL
#15
MLM15$Notes <- NULL
MLM15 <- MLM15 %>% drop_na()
MLM15$Y15 <- MLM15$`Age Adjusted Rate`
MLM15$Deaths <- NULL
MLM15$Population <- NULL
MLM15$`Crude Rate` <- NULL
MLM15$`Age Adjusted Rate` <- NULL
#16
MLM16$Notes <- NULL
MLM16 <- MLM16 %>% drop_na()
MLM16$Y16 <- MLM16$`Age Adjusted Rate`
MLM16$Deaths <- NULL
MLM16$Population <- NULL
MLM16$`Crude Rate` <- NULL
MLM16$`Age Adjusted Rate` <- NULL
#example of cleaned data layout
head(MLM16)# A tibble: 6 × 3
State `State Code` Y16
<chr> <chr> <dbl>
1 Alabama 01 502.
2 Alaska 02 379.
3 Arizona 04 355.
4 Arkansas 05 485.
5 California 06 269.
6 Colorado 08 292.6. Import Opioid data
#05
OP05 <- read_delim("OP05.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP05 <- OP05[-(which(OP05$State %in% "North Dakota")),]
OP05$`Age Adjusted Rate` <- replace(OP05$`Age Adjusted Rate`,
OP05$`Age Adjusted Rate` == "Unreliable", "NA")
OP05$`Age Adjusted Rate` <- OP05$`Age Adjusted Rate` %>% as.numeric()
OP05$Deaths <- OP05$Deaths %>% as.numeric()
OP05$Population <- OP05$Population %>% as.numeric()
OP05$Calculated <- round((OP05$Deaths/OP05$Population)*100000,
digits = 1)
OP05$`Age Adjusted Rate` <- ifelse(is.na(OP05$`Age Adjusted Rate`),
OP05$Calculated,
OP05$`Age Adjusted Rate`)
#06
OP06 <- read_delim("OP06.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP06 <- OP06[-(which(OP06$State %in% "North Dakota")),]
OP06$`Age Adjusted Rate` <- replace(OP06$`Age Adjusted Rate`,
OP06$`Age Adjusted Rate` == "Unreliable", "NA")
OP06$`Age Adjusted Rate` <- OP06$`Age Adjusted Rate` %>% as.numeric()
OP06$Deaths <- OP06$Deaths %>% as.numeric()
OP06$Population <- OP06$Population %>% as.numeric()
OP06$Calculated <- round((OP06$Deaths/OP06$Population)*100000,
digits = 1)
OP06$`Age Adjusted Rate` <- ifelse(is.na(OP06$`Age Adjusted Rate`),
OP06$Calculated,
OP06$`Age Adjusted Rate`)
#07
OP07 <- read_delim("OP07.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP07 <- OP07[-(which(OP07$State %in% "North Dakota")),]
OP07$`Age Adjusted Rate` <- replace(OP07$`Age Adjusted Rate`,
OP07$`Age Adjusted Rate` == "Unreliable", "NA")
OP07$`Age Adjusted Rate` <- OP07$`Age Adjusted Rate` %>% as.numeric()
OP07$Deaths <- OP07$Deaths %>% as.numeric()
OP07$Population <- OP07$Population %>% as.numeric()
OP07$Calculated <- round((OP07$Deaths/OP07$Population)*100000,
digits = 1)
OP07$`Age Adjusted Rate` <- ifelse(is.na(OP07$`Age Adjusted Rate`),
OP07$Calculated,
OP07$`Age Adjusted Rate`)
#08
OP08 <- read_delim("OP08.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP08 <- OP08[-(which(OP08$State %in% "North Dakota")),]
OP08$`Age Adjusted Rate` <- replace(OP08$`Age Adjusted Rate`, OP08$`Age Adjusted Rate` == "Unreliable", "NA")
OP08$`Age Adjusted Rate` <- OP08$`Age Adjusted Rate` %>% as.numeric()
OP08$Deaths <- OP08$Deaths %>% as.numeric()
OP08$Population <- OP08$Population %>% as.numeric()
OP08$Calculated <- round((OP08$Deaths/OP08$Population)*100000,digits = 1)
OP08$`Age Adjusted Rate` <- ifelse(is.na(OP08$`Age Adjusted Rate`), OP08$Calculated, OP08$`Age Adjusted Rate`)
#09
OP09 <- read_delim("OP09.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP09 <- OP09[-(which(OP09$State %in% "North Dakota")),]
OP09$`Age Adjusted Rate` <- replace(OP09$`Age Adjusted Rate`, OP09$`Age Adjusted Rate` == "Unreliable", "NA")
OP09$`Age Adjusted Rate` <- OP09$`Age Adjusted Rate` %>% as.numeric()
OP09$Deaths <- OP09$Deaths %>% as.numeric()
OP09$Population <- OP09$Population %>% as.numeric()
OP09$Calculated <- round((OP09$Deaths/OP09$Population)*100000,digits = 1)
OP09$`Age Adjusted Rate` <- ifelse(is.na(OP09$`Age Adjusted Rate`), OP09$Calculated, OP09$`Age Adjusted Rate`)
#10
OP10 <- read_delim("OP10.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP10 <- OP10[-(which(OP10$State %in% "North Dakota")),]
OP10$`Age Adjusted Rate` <- replace(OP10$`Age Adjusted Rate`, OP10$`Age Adjusted Rate` == "Unreliable", "NA")
OP10$`Age Adjusted Rate` <- OP10$`Age Adjusted Rate` %>% as.numeric()
OP10$Deaths <- OP10$Deaths %>% as.numeric()
OP10$Population <- OP10$Population %>% as.numeric()
OP10$Calculated <- round((OP10$Deaths/OP10$Population)*100000,digits = 1)
OP10$`Age Adjusted Rate` <- ifelse(is.na(OP10$`Age Adjusted Rate`), OP10$Calculated, OP10$`Age Adjusted Rate`)
#11
OP11 <- read_delim("OP11.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP11 <- OP11[-(which(OP11$State %in% "North Dakota")),]
OP11$`Age Adjusted Rate` <- replace(OP11$`Age Adjusted Rate`, OP11$`Age Adjusted Rate` == "Unreliable", "NA")
OP11$`Age Adjusted Rate` <- OP11$`Age Adjusted Rate` %>% as.numeric()
OP11$Deaths <- OP11$Deaths %>% as.numeric()
OP11$Population <- OP11$Population %>% as.numeric()
OP11$Calculated <- round((OP11$Deaths/OP11$Population)*100000,digits = 1)
OP11$`Age Adjusted Rate` <- ifelse(is.na(OP11$`Age Adjusted Rate`), OP11$Calculated, OP11$`Age Adjusted Rate`)
#12
OP12 <- read_delim("OP12.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP12 <- OP12[-(which(OP12$State %in% "North Dakota")),]
OP12$`Age Adjusted Rate` <- replace(OP12$`Age Adjusted Rate`, OP12$`Age Adjusted Rate` == "Unreliable", "NA")
OP12$`Age Adjusted Rate` <- OP12$`Age Adjusted Rate` %>% as.numeric()
OP12$Deaths <- OP12$Deaths %>% as.numeric()
OP12$Population <- OP12$Population %>% as.numeric()
OP12$Calculated <- round((OP12$Deaths/OP12$Population)*100000,digits = 1)
OP12$`Age Adjusted Rate` <- ifelse(is.na(OP12$`Age Adjusted Rate`), OP12$Calculated, OP12$`Age Adjusted Rate`)
#13
OP13 <- read_delim("OP13.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP13 <- OP13[-(which(OP13$State %in% "North Dakota")),]
OP13$`Age Adjusted Rate` <- replace(OP13$`Age Adjusted Rate`, OP13$`Age Adjusted Rate` == "Unreliable", "NA")
OP13$`Age Adjusted Rate` <- OP13$`Age Adjusted Rate` %>% as.numeric()
OP13$Deaths <- OP13$Deaths %>% as.numeric()
OP13$Population <- OP13$Population %>% as.numeric()
OP13$Calculated <- round((OP13$Deaths/OP13$Population)*100000,digits = 1)
OP13$`Age Adjusted Rate` <- ifelse(is.na(OP13$`Age Adjusted Rate`), OP13$Calculated, OP13$`Age Adjusted Rate`)
#14
OP14 <- read_delim("OP14.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP14 <- OP14[-(which(OP14$State %in% "North Dakota")),]
OP14$`Age Adjusted Rate` <- replace(OP14$`Age Adjusted Rate`, OP14$`Age Adjusted Rate` == "Unreliable", "NA")
OP14$`Age Adjusted Rate` <- OP14$`Age Adjusted Rate` %>% as.numeric()
OP14$Deaths <- OP14$Deaths %>% as.numeric()
OP14$Population <- OP14$Population %>% as.numeric()
OP14$Calculated <- round((OP14$Deaths/OP14$Population)*100000,digits = 1)
OP14$`Age Adjusted Rate` <- ifelse(is.na(OP14$`Age Adjusted Rate`), OP14$Calculated, OP14$`Age Adjusted Rate`)
#15
OP15 <- read_delim("OP15.txt",
delim = "\t", escape_double = FALSE,
trim_ws = TRUE)
OP15 <- OP15[-(which(OP15$State %in% "North Dakota")),]
OP15$`Age Adjusted Rate` <- replace(OP15$`Age Adjusted Rate`, OP15$`Age Adjusted Rate` == "Unreliable", "NA")
OP15$`Age Adjusted Rate` <- OP15$`Age Adjusted Rate` %>% as.numeric()
OP15$Deaths <- OP15$Deaths %>% as.numeric()
OP15$Population <- OP15$Population %>% as.numeric()
OP15$Calculated <- round((OP15$Deaths/OP15$Population)*100000,digits = 1)
OP15$`Age Adjusted Rate` <- ifelse(is.na(OP15$`Age Adjusted Rate`), OP15$Calculated, OP15$`Age Adjusted Rate`)7. Clean Opioid data
#05
OP05$Notes <- NULL
OP05 <- OP05 %>% drop_na()
OP05$Y05 <- OP05$`Age Adjusted Rate`
OP05$Deaths <- NULL
OP05$Population <- NULL
OP05$`Crude Rate` <- NULL
OP05$`Age Adjusted Rate` <- NULL
OP05$Calculated <- NULL
#06
OP06$Notes <- NULL
OP06 <- OP06 %>% drop_na()
OP06$Y06 <- OP06$`Age Adjusted Rate`
OP06$Deaths <- NULL
OP06$Population <- NULL
OP06$`Crude Rate` <- NULL
OP06$`Age Adjusted Rate` <- NULL
OP06$Calculated <- NULL
#07
OP07$Notes <- NULL
OP07 <- OP07 %>% drop_na()
OP07$Y07 <- OP07$`Age Adjusted Rate`
OP07$Deaths <- NULL
OP07$Population <- NULL
OP07$`Crude Rate` <- NULL
OP07$`Age Adjusted Rate` <- NULL
OP07$Calculated <- NULL
#08
OP08$Notes <- NULL
OP08 <- OP08 %>% drop_na()
OP08$Y08 <- OP08$`Age Adjusted Rate`
OP08$Deaths <- NULL
OP08$Population <- NULL
OP08$`Crude Rate` <- NULL
OP08$`Age Adjusted Rate` <- NULL
OP08$Calculated <- NULL
#09
OP09$Notes <- NULL
OP09 <- OP09 %>% drop_na()
OP09$Y09 <- OP09$`Age Adjusted Rate`
OP09$Deaths <- NULL
OP09$Population <- NULL
OP09$`Crude Rate` <- NULL
OP09$`Age Adjusted Rate` <- NULL
OP09$Calculated <- NULL
#10
OP10$Notes <- NULL
OP10 <- OP10 %>% drop_na()
OP10$Y10 <- OP10$`Age Adjusted Rate`
OP10$Deaths <- NULL
OP10$Population <- NULL
OP10$`Crude Rate` <- NULL
OP10$`Age Adjusted Rate` <- NULL
OP10$Calculated <- NULL
#11
OP11$Notes <- NULL
OP11 <- OP11 %>% drop_na()
OP11$Y11 <- OP11$`Age Adjusted Rate`
OP11$Deaths <- NULL
OP11$Population <- NULL
OP11$`Crude Rate` <- NULL
OP11$`Age Adjusted Rate` <- NULL
OP11$Calculated <- NULL
#12
OP12$Notes <- NULL
OP12 <- OP12 %>% drop_na()
OP12$Y12 <- OP12$`Age Adjusted Rate`
OP12$Deaths <- NULL
OP12$Population <- NULL
OP12$`Crude Rate` <- NULL
OP12$`Age Adjusted Rate` <- NULL
OP12$Calculated <- NULL
#13
OP13$Notes <- NULL
OP13 <- OP13 %>% drop_na()
OP13$Y13 <- OP13$`Age Adjusted Rate`
OP13$Deaths <- NULL
OP13$Population <- NULL
OP13$`Crude Rate` <- NULL
OP13$`Age Adjusted Rate` <- NULL
OP13$Calculated <- NULL
#14
OP14$Notes <- NULL
OP14 <- OP14 %>% drop_na()
OP14$Y14 <- OP14$`Age Adjusted Rate`
OP14$Deaths <- NULL
OP14$Population <- NULL
OP14$`Crude Rate` <- NULL
OP14$`Age Adjusted Rate` <- NULL
OP14$Calculated <- NULL
#15
OP15$Notes <- NULL
OP15 <- OP15 %>% drop_na()
OP15$Y15 <- OP15$`Age Adjusted Rate`
OP15$Deaths <- NULL
OP15$Population <- NULL
OP15$`Crude Rate` <- NULL
OP15$`Age Adjusted Rate` <- NULL
OP15$Calculated <- NULL8. Join data sets
DODlist_df = list(DOD05, DOD06, DOD07, DOD08, DOD09, DOD10, DOD11, DOD12, DOD13, DOD14, DOD15, DOD16)
DODtestdf <- DODlist_df %>% reduce(inner_join, by=c('State', 'State Code'))
MLMlist_df = list(MLM05, MLM06, MLM07, MLM08, MLM09, MLM10, MLM11, MLM12, MLM13, MLM14, MLM15, MLM16)
MLMtestdf <- MLMlist_df %>% reduce(inner_join, by=c('State', 'State Code'))
OPlist_df =list(OP05, OP06, OP07, OP08, OP09, OP10, OP11, OP12, OP13, OP14, OP15)
OPtestdf <- OPlist_df %>% reduce(inner_join,by=c('State', 'State Code'))
fulllist <- list(DODtestdf, MLMtestdf, OPtestdf)
fulldataset <- fulllist %>% reduce(inner_join, by=c('State', 'State Code'))9. create long deaths of despair data frame
longDODdata <- DODtestdf %>% pivot_longer(cols = c(Y05, Y06, Y07, Y08, Y09, Y10, Y11, Y12, Y13, Y14, Y15, Y16),
names_to = 'year',
values_to = 'DDR')10. create long midlife mortality data frame
longMLMdata <- MLMtestdf %>% pivot_longer(cols=c(Y05, Y06, Y07, Y08, Y09, Y10, Y11, Y12, Y13, Y14, Y15, Y16),
names_to = 'year',
values_to = 'MLDR')11. create long opioid data frame
longOPdata <- OPtestdf %>% pivot_longer(cols = c(Y05, Y06, Y07, Y08, Y09, Y10, Y11, Y12, Y13, Y14, Y15),
names_to = 'year',
values_to = 'OPDR')12. Long format for all mortality indicators
fulllonglist <- list(longDODdata, longMLMdata, longOPdata)
fulllongdataset <- fulllonglist %>%
reduce(inner_join, by=c('State', 'State Code', 'year'))13. read in non marital fertility rates and clean
nmfr <- read_csv("nmfr.csv")
nmfr$...1 <- NULL
nmfr$Pop <- NULL
nmfr$Births <- NULL
nmfr$STATEFIP <- nmfr$STATEFIP %>% str_pad(2,pad = "0")
nmfr$year5 <- nmfr$YEAR-5
nmfr$year4 <- nmfr$YEAR-4
nmfr$year3 <- nmfr$YEAR-3
nmfr$year2 <- nmfr$YEAR-2
nmfr$year1 <- nmfr$YEAR-1
fulllongdataset$STATEFIP <- fulllongdataset$`State Code`
fulllongdataset$`State Code` <- NULL
fulllongdataset <- fulllongdataset %>%
mutate(year = recode(year, Y05 = '2005', Y06='2006', Y07='2007',Y08='2008',Y09='2009',Y10='2010',Y11='2011',Y12='2012',Y13='2013',Y14='2014',Y15='2015'))
nmfr <- nmfr %>% drop_na()14. set up five year lag
#with a five year lag
fiveyeartestmerge <- merge(fulllongdataset, nmfr, by.x=c('STATEFIP', 'year'), by.y=c('STATEFIP', 'year5'))15. set up four year lag
fouryeartestmerge <- merge(fulllongdataset, nmfr, by.x=c('STATEFIP', 'year'), by.y=c('STATEFIP', 'year4'))16. set up three year lag
threeyeartestmerge <- merge(fulllongdataset, nmfr, by.x=c('STATEFIP', 'year'), by.y=c('STATEFIP', 'year3'))17. set up two year lag
twoyeartestmerge <- merge(fulllongdataset, nmfr, by.x=c('STATEFIP', 'year'), by.y=c('STATEFIP', 'year2'))18. set up one year lag
oneyeartestmerge <- merge(fulllongdataset, nmfr, by.x=c('STATEFIP', 'year'), by.y=c('STATEFIP', 'year1'))19. graphs for deaths of despair
fulllongdataset %>% ggplot(aes(x=year, y=DDR, group=State))+
geom_line(aes(color=State))+
geom_point(aes(color=State))+
ggtitle("Mortality Rates for \n Deaths of Despair")+
xlab("Year")+
ylab("Mortality Rate per 100,000")+
labs(caption = "Mortality rates are age standardized \n Source: CDC WONDER")
fulllongdataset %>% ggplot(aes(x=year, y=DDR, group=State))+
geom_line()+
geom_point()+
facet_wrap(~ State)+
ggtitle("Yearly mortality rates for deaths \n classified as deaths of despair")+
xlab("Year")+
ylab("Mortality Rate per 100,000")+
labs(caption = "Mortality rates are age standardized \n Source: CDC WONDER")
20. graphs for midlife mortality
fulllongdataset %>% ggplot(aes(x=year, y=MLDR, group=State))+
geom_line(aes(color=State))+
geom_point(aes(color=State))+
ggtitle("Mortality rates \n for Midlife Mortality")+
xlab("Year")+
ylab("Mortality Rate per 100,000")+
labs(caption = "Mortality rates are age standardized \n Source: CDC WONDER")
fulllongdataset %>% ggplot(aes(x=year, y=MLDR, group=State))+
geom_line()+
geom_point()+
facet_wrap(~State)+
ggtitle("Yearly mortality rates for midlife mortality")+
xlab("Year")+
ylab("Mortality Rate per 100,000")+
labs(caption = "Mortality rates are age standardized \n Source: CDC WONDER")
21. graphs for opioid deaths
fulllongdataset %>% ggplot(aes(x=year, y=OPDR, group=State))+
geom_line(aes(color=State))+
geom_point(aes(color=State))+
ggtitle("Mortality Rates for \n Deaths with Opioids")+
xlab("Year")+
ylab("Mortality Rate per 100,000")+
labs(caption = "Mortality rates are age standardized \n Source: CDC WONDER")
fulllongdataset %>% ggplot(aes(x=year, y=OPDR, group=State))+
geom_line()+
geom_point()+
facet_wrap(~ State)+
ggtitle("Yearly mortality rates for deaths with opioids")+
xlab("Year")+
ylab("Mortality Rate per 100,000")+
labs(caption = "Mortality rates are age standardized \n Source: CDC WONDER")
22. graphs for nonmarital fertility
fiveyeartestmerge %>% ggplot(aes(x=YEAR, y=nmfr, group=State))+
geom_line(aes(color=State))+
geom_point(aes(color=State))+
ggtitle("Nonmarital Fertility Rates")+
xlab("Year")+
ylab("Nonmarital fertiltiy rate")+
labs(caption = "Source: ACS")
oneyeartestmerge %>% ggplot(aes(x=YEAR, y=nmfr, group=State))+geom_line()+geom_point()+facet_wrap(~State)+
ggtitle("Yearly nonmarital fertility rates")+
xlab("Year")+
ylab("Nonmarital fertiltiy rate")+
labs(caption = "Source: ACS")
23. set up data frames for maps
states2<-states%>%
filter(substr(GEOID, 1, 2) != "02") %>%
filter(substr(GEOID, 1, 2) != "15") %>%
filter(substr(GEOID, 1, 2) != "66") %>%
filter(substr(GEOID, 1, 2) != "72")
states2$STATEFIP<-states2$GEOID
full<-merge(fiveyeartestmerge,states2, by="STATEFIP")24. Maps for midlife mortality
We exclude North Dakota due to suppressions for certain causes of death from CDC WONDER. California is missing data for some years due to state privacy concerns with data sharing.
full2020<-full%>%
filter(YEAR=="2020")
full2020 <- st_sf(full2020)
full2015 <- full %>%
filter(YEAR=="2015")
full2015 <- st_sf(full2015)
full2010<-full%>%
filter(YEAR=="2010")
full2010 <- st_sf(full2010)
ggplot(data=full2010, aes(fill = MLDR))+ geom_sf()+
ggtitle("Mortality Rates for Midlife Mortality - 2010")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n Midlife mortality defined as ages 25-64")
ggplot(data=full2015, aes(fill = MLDR))+ geom_sf()+
ggtitle("Mortality Rates for Midlife Mortality - 2015")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n Midlife mortality defined as ages 25-64")
ggplot(data=full2020, aes(fill = MLDR))+ geom_sf()+
ggtitle("Mortality Rates for Midlife Mortality - 2020")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n Midlife mortality defined as ages 25-64")
25. Map for deaths of despair
ggplot(data=full2010, aes(fill = DDR))+ geom_sf()+
ggtitle("Mortality Rates for deaths of despair - 2010")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n Deaths of despair follow Case and Deaton's classifications")
ggplot(data=full2015, aes(fill = DDR))+ geom_sf()+
ggtitle("Mortality Rates for deaths of despair - 2015")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n Deaths of despair follow Case and Deaton's classifications")
ggplot(data=full2020, aes(fill = DDR))+ geom_sf()+
ggtitle("Mortality Rates for deaths of despair - 2020")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n Deaths of despair follow Case and Deaton's classifications")
26. maps for opioids
ggplot(data=full2010, aes(fill = OPDR))+ geom_sf()+
ggtitle("Mortality Rates from Opioids - 2010")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n ")
ggplot(data=full2015, aes(fill = OPDR))+ geom_sf()+
ggtitle("Mortality Rates from Opioids - 2015")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n ")
ggplot(data=full2020, aes(fill = OPDR))+ geom_sf()+
ggtitle("Mortality Rates from Opioids - 2020")+
ggthemes::theme_map()+
labs(caption = "Source: CDC WONDER \n ")
Nonmarital fertility maps
ggplot(data=full2010, aes(fill = nmfr))+ geom_sf()+
ggtitle("Nonmarital Fertility Rates - 2010")+
ggthemes::theme_map()+
labs(caption = "Source: ACS")+
scale_fill_distiller(palette = "RdPu", direction = 1)
ggplot(data=full2015, aes(fill = nmfr))+ geom_sf()+
ggtitle("Nonmarital Fertility Rates - 2015")+
ggthemes::theme_map()+
labs(caption = "Source: ACS")+
scale_fill_distiller(palette = "RdPu", direction = 1)
ggplot(data=full2020, aes(fill = nmfr))+ geom_sf()+
ggtitle("Nonmarital Fertility Rates - 2020")+
ggthemes::theme_map()+
labs(caption = "Source: ACS")+
scale_fill_distiller(palette = "RdPu", direction = 1)
27. Bivariate plots with five year lag for midlife mortality and nonmarital ferility
fiveyeartestmerge2020 <- fiveyeartestmerge %>%
filter(YEAR==2020)
fiveyeartestmerge2015 <- fiveyeartestmerge %>%
filter(YEAR==2015)
fiveyeartestmerge2010 <- fiveyeartestmerge %>%
filter(YEAR==2010)
ggplot(data=fiveyeartestmerge2020, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2020 (Five Year Lag)")+
labs(caption = "2020 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=fiveyeartestmerge2015, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2015 (Five Year Lag)")+
labs(caption = "2015 Ferility and 2010 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility Rate")
ggplot(data=fiveyeartestmerge2010, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2010 (Five Year Lag)")+
labs(caption = "2010 Ferility and 2005 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
28. Bivariate plots with three year lag for midlife mortality and nonmarital ferility
threeyeartestmerge2018 <- threeyeartestmerge %>%
filter(YEAR==2018)
threeyeartestmerge2014 <- threeyeartestmerge %>%
filter(YEAR==2014)
threeyeartestmerge2010 <- threeyeartestmerge %>%
filter(YEAR==2010)
ggplot(data=threeyeartestmerge2018, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2018 (Three Year Lag)")+
labs(caption = "2018 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=threeyeartestmerge2014, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2014 (Three Year Lag)")+
labs(caption = "2014 Ferility and 2012 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=threeyeartestmerge2010, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2010 (Three Year Lag)")+
labs(caption = "2010 Ferility and 2007 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
29. Bivariate plots with one year lag for midlife mortality and nonmarital ferility
oneyearyeartestmerge2016 <- oneyeartestmerge %>%
filter(YEAR==2016)
oneyeartestmerge2013 <- oneyeartestmerge %>%
filter(YEAR==2013)
oneyeartestmerge2010 <- oneyeartestmerge %>%
filter(YEAR==2010)
ggplot(data=oneyearyeartestmerge2016, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2016 (One Year Lag)")+
labs(caption = "2016 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=oneyeartestmerge2013, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2013 (One Year Lag)")+
labs(caption = "2013 Ferility and 2012 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=oneyeartestmerge2010, aes(x=MLDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Midlife Mortality and Nonmarital Fertility - 2010 (One Year Lag)")+
labs(caption = "2010 Ferility and 2009 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Midlife Mortality Rate")+
ylab("Nonmarital Fertility rate")
30. Bivariate plots with five year lag for deaths of despair and nonmarital fertility
ggplot(data=fiveyeartestmerge2020, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2020 (Five Year Lag)")+
labs(caption = "2020 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=fiveyeartestmerge2015, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2015 (Five Year Lag)")+
labs(caption = "2015 Ferility and 2010 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=fiveyeartestmerge2010, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2010 (Five Year Lag)")+
labs(caption = "2010 Ferility and 2005 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
31. Bivariate plots with three year lag for deaths of despair and nonmarital fertility
ggplot(data=threeyeartestmerge2018, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2018 (Three Year Lag)")+
labs(caption = "2018 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=threeyeartestmerge2014, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2014 (Three Year Lag)")+
labs(caption = "2014 Ferility and 2011 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=threeyeartestmerge2010, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2010 (Three Year Lag)")+
labs(caption = "2010 Ferility and 2007 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
32. Bivariate plots with one year lags for deaths of despair deaths and nonmarital ferility
ggplot(data=oneyearyeartestmerge2016, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2016 (One Year Lag)")+
labs(caption = "2016 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=oneyeartestmerge2013, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2013 (One Year Lag)")+
labs(caption = "2013 Ferility and 2012 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=oneyeartestmerge2010, aes(x=DDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Deaths of Despair and Nonmarital Fertility - 2010 (One Year Lag)")+
labs(caption = "2010 Ferility and 2009 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Deaths of Despair Mortality Rate")+
ylab("Nonmarital Fertility rate")
33. Bivariate plots with five year lags for opioid deaths and nonmarital ferility
ggplot(data=fiveyeartestmerge2020, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2020 (Five Year Lag)")+
labs(caption = "2020 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=fiveyeartestmerge2015, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2015 (Five Year Lag)")+
labs(caption = "2015 Ferility and 2010 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=fiveyeartestmerge2010, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2010 (Five Year Lag)")+
labs(caption = "2010 Ferility and 2005 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
34. Bivariate plots with three year lags for opioid deaths and nonmarital ferility
ggplot(data=threeyeartestmerge2018, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2018 (Three Year Lag)")+
labs(caption = "2018 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=threeyeartestmerge2014, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2014 (Three Year Lag)")+
labs(caption = "2014 Ferility and 2011 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=threeyeartestmerge2010, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2010 (Three Year Lag)")+
labs(caption = "2010 Ferility and 2007 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
35. Bivariate plots with one year lags for opioid deaths and nonmarital ferility
ggplot(data=oneyearyeartestmerge2016, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2016 (One Year Lag)")+
labs(caption = "2016 Ferility and 2015 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=oneyeartestmerge2013, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2013 (One Year Lag)")+
labs(caption = "2013 Ferility and 2012 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
ggplot(data=oneyeartestmerge2010, aes(x=OPDR, y=nmfr)) +
geom_point() +
geom_smooth(method=lm)+
ggtitle("Opioid Mortality and Nonmarital Fertility - 2010 (One Year Lag)")+
labs(caption = "2010 Ferility and 2009 Mortality Rates \n Sources: ACS and CDC WONDER")+
xlab("Opioid related Mortality Rate")+
ylab("Nonmarital Fertility rate")
Now, we show the correlation between the three different mortality indicators with nonmarital fertility with different lag schemes. Some of the coefficients are reasonably strong (such as midlife mortality and nonmaritial ferility which is consistent with the previous plots).
36. correlation between different indicators with five year lag
#mid life mortality and nonmarital fertility
cor(fiveyeartestmerge$MLDR, fiveyeartestmerge$nmfr)[1] 0.6495454#deaths of despair and nonmarital fertility
cor(fiveyeartestmerge$DDR, fiveyeartestmerge$nmfr)[1] 0.05213902#opioid deaths and nonmarital fertility
cor(fiveyeartestmerge$OPDR, fiveyeartestmerge$nmfr)[1] -0.16627937. correlation between different indicators with four year lag
#mid life mortality and nonmarital fertility
cor(fouryeartestmerge$MLDR, fouryeartestmerge$nmfr)[1] 0.6507215#deaths of despair and nonmarital fertility
cor(fouryeartestmerge$DDR, fouryeartestmerge$nmfr)[1] 0.08268375#opioid deaths and nonmarital fertility
cor(fouryeartestmerge$OPDR, fouryeartestmerge$nmfr)[1] -0.136853138. correlation between different indicators with three year lag
#mid life mortality and nonmarital fertility
cor(threeyeartestmerge$MLDR, threeyeartestmerge$nmfr)[1] 0.6513654#deaths of despair and nonmarital fertility
cor(threeyeartestmerge$DDR, threeyeartestmerge$nmfr)[1] 0.09573672#opioid deaths and nonmarital fertility
cor(threeyeartestmerge$OPDR, threeyeartestmerge$nmfr)[1] -0.122698339. correlation between different indicators with two year lag
#mid life mortality and nonmarital fertility
cor(twoyeartestmerge$MLDR, twoyeartestmerge$nmfr)[1] 0.6583056#deaths of despair and nonmarital fertility
cor(twoyeartestmerge$DDR, twoyeartestmerge$nmfr)[1] 0.1138378#opioid deaths and nonmarital fertility
cor(twoyeartestmerge$OPDR, twoyeartestmerge$nmfr)[1] -0.110938740. correlation between different indicators with one year lag
#mid life mortality and nonmarital fertility
cor(oneyeartestmerge$MLDR, oneyeartestmerge$nmfr)[1] 0.674302#deaths of despair and nonmarital fertility
cor(oneyeartestmerge$DDR, oneyeartestmerge$nmfr)[1] 0.125499#opioid deaths and nonmarital fertility
cor(oneyeartestmerge$OPDR, oneyeartestmerge$nmfr)[1] -0.1156244Next, we present fixed and random effect models for the three different mortality indicators with nonmarital fertility (we use lag times of five, four, three, two, and one years). We will interpret an early output as an example. We also run the Hausman test for every set of models to test whether fixed or random effects is preferable.
41. fixed effect models with states included with five year lags
MLmodel5 <- lm(nmfr ~ MLDR + State -1, data=fiveyeartestmerge)
summary(MLmodel5)
Call:
lm(formula = nmfr ~ MLDR + State - 1, data = fiveyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.403 -2.687 0.320 2.665 18.012
Coefficients:
Estimate Std. Error t value Pr(>|t|)
MLDR 0.03589 0.01489 2.410 0.01631 *
StateAlabama 39.85560 7.29915 5.460 7.53e-08 ***
StateAlaska 43.20804 5.31573 8.128 3.50e-15 ***
StateArizona 49.66729 5.30248 9.367 < 2e-16 ***
StateArkansas 51.32047 7.06274 7.266 1.44e-12 ***
StateCalifornia 38.91880 4.56460 8.526 < 2e-16 ***
StateColorado 24.78962 4.43803 5.586 3.84e-08 ***
StateConnecticut 25.49018 4.18153 6.096 2.19e-09 ***
StateDelaware 35.91660 5.59289 6.422 3.16e-10 ***
StateDistrict of Columbia 19.81661 7.29424 2.717 0.00682 **
StateFlorida 41.37817 5.49338 7.532 2.38e-13 ***
StateGeorgia 43.98924 5.89061 7.468 3.70e-13 ***
StateHawaii 40.64329 4.48653 9.059 < 2e-16 ***
StateIdaho 35.35163 4.67320 7.565 1.90e-13 ***
StateIllinois 38.48395 4.94521 7.782 4.18e-14 ***
StateIndiana 44.49585 5.75567 7.731 5.99e-14 ***
StateIowa 39.05491 4.65229 8.395 4.92e-16 ***
StateKansas 44.09800 5.17540 8.521 < 2e-16 ***
StateKentucky 46.74437 6.85299 6.821 2.64e-11 ***
StateLouisiana 49.08901 7.16083 6.855 2.12e-11 ***
StateMaine 30.28824 4.80598 6.302 6.49e-10 ***
StateMaryland 31.86271 5.09558 6.253 8.69e-10 ***
StateMassachusetts 29.73916 4.27632 6.954 1.12e-11 ***
StateMichigan 34.48460 5.53953 6.225 1.03e-09 ***
StateMinnesota 33.74476 3.96930 8.501 < 2e-16 ***
StateMississippi 50.12110 7.77148 6.449 2.67e-10 ***
StateMissouri 40.89313 5.92337 6.904 1.56e-11 ***
StateMontana 35.07275 5.31274 6.602 1.05e-10 ***
StateNebraska 39.60828 4.55262 8.700 < 2e-16 ***
StateNevada 46.70467 5.74599 8.128 3.50e-15 ***
StateNew Hampshire 24.29951 4.28888 5.666 2.49e-08 ***
StateNew Jersey 27.73755 4.46942 6.206 1.15e-09 ***
StateNew Mexico 45.88552 6.07038 7.559 1.98e-13 ***
StateNew York 32.87743 4.44071 7.404 5.72e-13 ***
StateNorth Carolina 38.48898 5.67763 6.779 3.45e-11 ***
StateOhio 42.22715 5.74821 7.346 8.45e-13 ***
StateOklahoma 54.02154 7.09331 7.616 1.34e-13 ***
StateOregon 30.43350 4.86446 6.256 8.53e-10 ***
StatePennsylvania 32.92890 5.28429 6.231 9.88e-10 ***
StateRhode Island 27.37241 4.68283 5.845 9.17e-09 ***
StateSouth Carolina 42.28369 6.56639 6.439 2.84e-10 ***
StateSouth Dakota 46.34850 4.87451 9.508 < 2e-16 ***
StateTennessee 43.54850 6.81599 6.389 3.85e-10 ***
StateTexas 51.55881 5.31339 9.704 < 2e-16 ***
StateUtah 25.79309 4.53382 5.689 2.19e-08 ***
StateVermont 26.44752 4.29701 6.155 1.55e-09 ***
StateVirginia 37.58935 4.84062 7.765 4.70e-14 ***
StateWashington 32.54779 4.46227 7.294 1.20e-12 ***
StateWest Virginia 41.99450 7.27953 5.769 1.41e-08 ***
StateWisconsin 34.72824 4.52730 7.671 9.11e-14 ***
StateWyoming 38.33204 5.41836 7.074 5.14e-12 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.937 on 496 degrees of freedom
Multiple R-squared: 0.9917, Adjusted R-squared: 0.9909
F-statistic: 1168 on 51 and 496 DF, p-value: < 2.2e-16Dmodel5 <- lm(nmfr ~ DDR + State -1, data=fiveyeartestmerge)
summary(Dmodel5)
Call:
lm(formula = nmfr ~ DDR + State - 1, data = fiveyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.981 -1.805 -0.137 1.770 17.570
Coefficients:
Estimate Std. Error t value Pr(>|t|)
DDR -0.57693 0.03707 -15.56 <2e-16 ***
StateAlabama 74.45374 1.65907 44.88 <2e-16 ***
StateAlaska 86.84995 2.35829 36.83 <2e-16 ***
StateArizona 88.05735 2.07931 42.35 <2e-16 ***
StateArkansas 86.41136 1.70632 50.64 <2e-16 ***
StateCalifornia 66.34853 1.81696 36.52 <2e-16 ***
StateColorado 59.43811 1.99965 29.72 <2e-16 ***
StateConnecticut 50.14939 1.57006 31.94 <2e-16 ***
StateDelaware 67.48703 1.71195 39.42 <2e-16 ***
StateDistrict of Columbia 54.22436 1.65138 32.84 <2e-16 ***
StateFlorida 74.38321 1.78905 41.58 <2e-16 ***
StateGeorgia 73.18199 1.57871 46.36 <2e-16 ***
StateHawaii 65.92269 1.56356 42.16 <2e-16 ***
StateIdaho 68.29214 1.88479 36.23 <2e-16 ***
StateIllinois 64.09462 1.53094 41.87 <2e-16 ***
StateIndiana 76.83998 1.72848 44.45 <2e-16 ***
StateIowa 64.84602 1.56712 41.38 <2e-16 ***
StateKansas 73.61959 1.66771 44.14 <2e-16 ***
StateKentucky 86.96641 1.97580 44.02 <2e-16 ***
StateLouisiana 84.61050 1.71477 49.34 <2e-16 ***
StateMaine 60.82803 1.75502 34.66 <2e-16 ***
StateMaryland 58.87560 1.57111 37.47 <2e-16 ***
StateMassachusetts 55.64944 1.61121 34.54 <2e-16 ***
StateMichigan 66.52997 1.73943 38.25 <2e-16 ***
StateMinnesota 57.57127 1.55877 36.93 <2e-16 ***
StateMississippi 85.04661 1.62350 52.38 <2e-16 ***
StateMissouri 74.34756 1.76009 42.24 <2e-16 ***
StateMontana 74.19663 2.11620 35.06 <2e-16 ***
StateNebraska 63.73993 1.51261 42.14 <2e-16 ***
StateNevada 86.93381 2.11730 41.06 <2e-16 ***
StateNew Hampshire 54.67395 1.80876 30.23 <2e-16 ***
StateNew Jersey 50.71271 1.47797 34.31 <2e-16 ***
StateNew Mexico 96.12718 2.62173 36.66 <2e-16 ***
StateNew York 55.54317 1.46957 37.80 <2e-16 ***
StateNorth Carolina 69.28515 1.66840 41.53 <2e-16 ***
StateOhio 75.63311 1.77805 42.54 <2e-16 ***
StateOklahoma 96.42038 2.05652 46.88 <2e-16 ***
StateOregon 65.44314 1.96314 33.34 <2e-16 ***
StatePennsylvania 64.10364 1.72895 37.08 <2e-16 ***
StateRhode Island 58.47549 1.79568 32.56 <2e-16 ***
StateSouth Carolina 76.45814 1.72019 44.45 <2e-16 ***
StateSouth Dakota 77.44474 1.77269 43.69 <2e-16 ***
StateTennessee 81.36622 1.86162 43.71 <2e-16 ***
StateTexas 78.85206 1.56044 50.53 <2e-16 ***
StateUtah 61.44680 2.03845 30.14 <2e-16 ***
StateVermont 55.61031 1.75141 31.75 <2e-16 ***
StateVirginia 63.10323 1.53741 41.05 <2e-16 ***
StateWashington 64.23727 1.85024 34.72 <2e-16 ***
StateWest Virginia 87.00248 2.16896 40.11 <2e-16 ***
StateWisconsin 63.12224 1.69025 37.34 <2e-16 ***
StateWyoming 79.34676 2.20214 36.03 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.07 on 496 degrees of freedom
Multiple R-squared: 0.9944, Adjusted R-squared: 0.9938
F-statistic: 1722 on 51 and 496 DF, p-value: < 2.2e-16OPmodel5 <- lm(nmfr ~ OPDR + State -1, data=fiveyeartestmerge)
summary(OPmodel5)
Call:
lm(formula = nmfr ~ OPDR + State - 1, data = fiveyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-49.106 -2.026 -0.162 2.229 18.246
Coefficients:
Estimate Std. Error t value Pr(>|t|)
OPDR -0.69909 0.06858 -10.19 <2e-16 ***
StateAlabama 59.83597 1.38801 43.11 <2e-16 ***
StateAlaska 61.41754 1.47971 41.51 <2e-16 ***
StateArizona 67.91343 1.48241 45.81 <2e-16 ***
StateArkansas 72.25640 1.42513 50.70 <2e-16 ***
StateCalifornia 52.29979 1.62723 32.14 <2e-16 ***
StateColorado 40.09671 1.45487 27.56 <2e-16 ***
StateConnecticut 40.97739 1.48589 27.58 <2e-16 ***
StateDelaware 55.60853 1.51168 36.79 <2e-16 ***
StateDistrict of Columbia 42.33340 1.45752 29.05 <2e-16 ***
StateFlorida 59.58183 1.46293 40.73 <2e-16 ***
StateGeorgia 61.35449 1.40716 43.60 <2e-16 ***
StateHawaii 54.05972 1.39746 38.68 <2e-16 ***
StateIdaho 49.33913 1.39960 35.25 <2e-16 ***
StateIllinois 55.05480 1.45399 37.87 <2e-16 ***
StateIndiana 61.50562 1.40669 43.72 <2e-16 ***
StateIowa 53.02735 1.40047 37.86 <2e-16 ***
StateKansas 59.44441 1.40165 42.41 <2e-16 ***
StateKentucky 71.79587 1.61884 44.35 <2e-16 ***
StateLouisiana 68.88761 1.39113 49.52 <2e-16 ***
StateMaine 48.03199 1.51303 31.75 <2e-16 ***
StateMaryland 51.29164 1.55612 32.96 <2e-16 ***
StateMassachusetts 47.61856 1.58215 30.10 <2e-16 ***
StateMichigan 52.77200 1.46179 36.10 <2e-16 ***
StateMinnesota 45.88613 1.39873 32.81 <2e-16 ***
StateMississippi 70.84330 1.38058 51.31 <2e-16 ***
StateMissouri 61.09819 1.49865 40.77 <2e-16 ***
StateMontana 51.74946 1.42773 36.25 <2e-16 ***
StateNebraska 51.87784 1.37409 37.75 <2e-16 ***
StateNevada 70.21091 1.68552 41.66 <2e-16 ***
StateNew Hampshire 42.94843 1.62019 26.51 <2e-16 ***
StateNew Jersey 41.66952 1.41085 29.54 <2e-16 ***
StateNew Mexico 69.97088 1.67238 41.84 <2e-16 ***
StateNew York 47.52406 1.43309 33.16 <2e-16 ***
StateNorth Carolina 57.89695 1.49118 38.83 <2e-16 ***
StateOhio 63.34297 1.55854 40.64 <2e-16 ***
StateOklahoma 79.60871 1.61581 49.27 <2e-16 ***
StateOregon 47.70892 1.48764 32.07 <2e-16 ***
StatePennsylvania 49.46424 1.42568 34.70 <2e-16 ***
StateRhode Island 47.73339 1.65873 28.78 <2e-16 ***
StateSouth Carolina 61.93654 1.42348 43.51 <2e-16 ***
StateSouth Dakota 60.19204 1.38610 43.42 <2e-16 ***
StateTennessee 66.69735 1.53001 43.59 <2e-16 ***
StateTexas 66.88346 1.39348 48.00 <2e-16 ***
StateUtah 46.51182 1.70109 27.34 <2e-16 ***
StateVermont 42.47984 1.49580 28.40 <2e-16 ***
StateVirginia 53.36173 1.43643 37.15 <2e-16 ***
StateWashington 49.36590 1.51086 32.67 <2e-16 ***
StateWest Virginia 75.33005 2.08973 36.05 <2e-16 ***
StateWisconsin 50.63838 1.46824 34.49 <2e-16 ***
StateWyoming 55.46437 1.43348 38.69 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.515 on 496 degrees of freedom
Multiple R-squared: 0.9931, Adjusted R-squared: 0.9924
F-statistic: 1398 on 51 and 496 DF, p-value: < 2.2e-16For the first fixed effects model we test midlife mortality and nonmarital fertility with a five year lag (for example 2020 ferility rates and 2015 mortality rates). The coefficient of 0.03589 means that an increase in state-level midlife mortality is associated with a slight increase in state-level nonmarital fertility. This result is significant (at the 5% level).
The second fixed effects model tests deaths of despair and nonmarital fertility (again with a five year lag). The coefficient here of -0.57693 means that an increase in state-level mortality rates for deaths classified as deaths of despair is associated with a decrease in state-level nonmarital fertility. This result is significant.
The third fixed effects model tests opioid deaths and nonmarital fertility (also with a five year lag). The coefficient of -0.69909 means that an increase in state-level mortality rates for deaths involving opioids is associated with a drecrease in state-level nonmarital fertility. This result is significant.
42. fixed and random effect models with five year lags and Hausman test
MLfixed5 <- plm(nmfr ~ MLDR, data=fiveyeartestmerge, index=c("State", "year"), model="within")
MLrandom5 <- plm(nmfr ~ MLDR, data=fiveyeartestmerge, index=c("State", "year"), model="random")
summary(MLfixed5)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ MLDR, data = fiveyeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-11, N = 547
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.40267 -2.68686 0.32018 2.66470 18.01230
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
MLDR 0.035892 0.014892 2.4101 0.01631 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 12230
Residual Sum of Squares: 12089
R-Squared: 0.011575
Adj. R-Squared: -0.088064
F-statistic: 5.80868 on 1 and 496 DF, p-value: 0.01631summary(MLrandom5)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ MLDR, data = fiveyeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-11, N = 547
Effects:
var std.dev share
idiosyncratic 24.373 4.937 0.353
individual 44.693 6.685 0.647
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.7474 0.7827 0.7827 0.7821 0.7827 0.7827
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-47.053 -2.974 -0.102 -0.002 2.839 20.474
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 25.854607 3.706839 6.9748 3.062e-12 ***
MLDR 0.070832 0.010249 6.9114 4.800e-12 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 14706
Residual Sum of Squares: 13507
R-Squared: 0.081489
Adj. R-Squared: 0.079804
Chisq: 47.767 on 1 DF, p-value: 4.8001e-12phtest(MLfixed5, MLrandom5)
Hausman Test
data: nmfr ~ MLDR
chisq = 10.456, df = 1, p-value = 0.001222
alternative hypothesis: one model is inconsistentDfixed5 <- plm(nmfr ~ DDR, data=fiveyeartestmerge, index=c("State", "year"), model="within")
Drandom5 <- plm(nmfr ~ DDR, data=fiveyeartestmerge, index=c("State", "year"), model="random")
summary(Dfixed5)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ DDR, data = fiveyeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-11, N = 547
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.98105 -1.80475 -0.13666 1.77028 17.56991
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
DDR -0.576928 0.037068 -15.564 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 12230
Residual Sum of Squares: 8217.3
R-Squared: 0.32813
Adj. R-Squared: 0.2604
F-statistic: 242.236 on 1 and 496 DF, p-value: < 2.22e-16summary(Drandom5)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ DDR, data = fiveyeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-11, N = 547
Effects:
var std.dev share
idiosyncratic 16.567 4.070 0.151
individual 92.989 9.643 0.849
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.8524 0.8738 0.8738 0.8734 0.8738 0.8738
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-49.562 -2.170 -0.412 0.001 1.945 18.742
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 69.042633 1.896186 36.411 < 2.2e-16 ***
DDR -0.535105 0.036897 -14.503 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 13066
Residual Sum of Squares: 9423.8
R-Squared: 0.27876
Adj. R-Squared: 0.27744
Chisq: 210.33 on 1 DF, p-value: < 2.22e-16phtest(Dfixed5, Drandom5)
Hausman Test
data: nmfr ~ DDR
chisq = 137.93, df = 1, p-value < 2.2e-16
alternative hypothesis: one model is inconsistentOPfixed5 <- plm(nmfr ~ OPDR, data=fiveyeartestmerge, index=c("State", "year"), model="within")
OPrandom5 <- plm(nmfr ~ OPDR, data=fiveyeartestmerge, index=c("State", "year"), model="random")
summary(OPfixed5)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ OPDR, data = fiveyeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-11, N = 547
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-49.10644 -2.02609 -0.16164 2.22936 18.24563
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
OPDR -0.69909 0.06858 -10.194 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 12230
Residual Sum of Squares: 10112
R-Squared: 0.17322
Adj. R-Squared: 0.08987
F-statistic: 103.914 on 1 and 496 DF, p-value: < 2.22e-16summary(OPrandom5)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ OPDR, data = fiveyeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-11, N = 547
Effects:
var std.dev share
idiosyncratic 20.387 4.515 0.174
individual 96.691 9.833 0.826
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.8398 0.8629 0.8629 0.8625 0.8629 0.8629
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-49.734 -2.441 -0.330 0.002 2.358 19.713
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 56.162816 1.509538 37.205 < 2.2e-16 ***
OPDR -0.682532 0.067454 -10.118 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 13217
Residual Sum of Squares: 11121
R-Squared: 0.15855
Adj. R-Squared: 0.15701
Chisq: 102.383 on 1 DF, p-value: < 2.22e-16phtest(OPfixed5, OPrandom5)
Hausman Test
data: nmfr ~ OPDR
chisq = 1.7909, df = 1, p-value = 0.1808
alternative hypothesis: one model is inconsistentThese previous models for both fixed and random effects were used to run the Hausman test to determine if fixed effects or random effects is preferable. For example, for the midlife mortality models the Hausman test is significant meaning that the models are different which means that the fixed effects model is what we should use.
43. fixed effect models with states included with four year lags
MLmodel4 <- lm(nmfr ~ MLDR + State -1, data=fouryeartestmerge)
summary(MLmodel4)
Call:
lm(formula = nmfr ~ MLDR + State - 1, data = fouryeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.914 -2.512 -0.038 2.379 17.189
Coefficients:
Estimate Std. Error t value Pr(>|t|)
MLDR 0.004122 0.016398 0.251 0.802
StateAlabama 54.871036 8.000732 6.858 2.33e-11 ***
StateAlaska 54.469538 5.831145 9.341 < 2e-16 ***
StateArizona 61.873592 5.765566 10.732 < 2e-16 ***
StateArkansas 66.597245 7.758935 8.583 < 2e-16 ***
StateCalifornia 47.936527 4.862567 9.858 < 2e-16 ***
StateColorado 34.327940 4.834162 7.101 4.90e-12 ***
StateConnecticut 34.120900 4.537748 7.519 3.04e-13 ***
StateDelaware 47.904139 6.112406 7.837 3.40e-14 ***
StateDistrict of Columbia 35.853042 7.833441 4.577 6.12e-06 ***
StateFlorida 53.419500 5.976426 8.938 < 2e-16 ***
StateGeorgia 57.227797 6.410920 8.927 < 2e-16 ***
StateHawaii 50.099609 4.877053 10.273 < 2e-16 ***
StateIdaho 45.439783 5.110734 8.891 < 2e-16 ***
StateIllinois 49.183018 5.371719 9.156 < 2e-16 ***
StateIndiana 56.830494 6.315621 8.998 < 2e-16 ***
StateIowa 48.648892 5.086754 9.564 < 2e-16 ***
StateKansas 55.158041 5.649431 9.763 < 2e-16 ***
StateKentucky 61.624130 7.540089 8.173 3.13e-15 ***
StateLouisiana 64.598144 7.793851 8.288 1.35e-15 ***
StateMaine 40.583802 5.232833 7.756 6.00e-14 ***
StateMaryland 42.607623 5.525427 7.711 8.17e-14 ***
StateMassachusetts 38.904710 4.640502 8.384 6.74e-16 ***
StateMichigan 46.346207 6.053129 7.657 1.19e-13 ***
StateMinnesota 42.077933 4.308197 9.767 < 2e-16 ***
StateMississippi 66.921913 8.500918 7.872 2.66e-14 ***
StateMissouri 53.581980 6.484443 8.263 1.63e-15 ***
StateMontana 46.900619 5.808487 8.074 6.34e-15 ***
StateNebraska 49.272079 4.972959 9.908 < 2e-16 ***
StateNevada 59.503631 6.236483 9.541 < 2e-16 ***
StateNew Hampshire 33.264904 4.666582 7.128 4.11e-12 ***
StateNew Jersey 37.079612 4.838683 7.663 1.14e-13 ***
StateNew Mexico 59.407680 6.676042 8.899 < 2e-16 ***
StateNew York 42.292714 4.807667 8.797 < 2e-16 ***
StateNorth Carolina 50.809271 6.188265 8.211 2.38e-15 ***
StateOhio 54.471333 6.296986 8.650 < 2e-16 ***
StateOklahoma 69.453265 7.786449 8.920 < 2e-16 ***
StateOregon 41.039814 5.300270 7.743 6.55e-14 ***
StatePennsylvania 43.900385 5.759232 7.623 1.50e-13 ***
StateRhode Island 37.128914 5.111832 7.263 1.69e-12 ***
StateSouth Carolina 56.620089 7.173265 7.893 2.29e-14 ***
StateSouth Dakota 57.291821 5.291935 10.826 < 2e-16 ***
StateTennessee 58.381822 7.463449 7.822 3.77e-14 ***
StateTexas 63.762663 5.789003 11.014 < 2e-16 ***
StateUtah 35.676343 4.952638 7.204 2.51e-12 ***
StateVermont 35.560147 4.711485 7.548 2.51e-13 ***
StateVirginia 47.906878 5.256875 9.113 < 2e-16 ***
StateWashington 42.116859 4.853965 8.677 < 2e-16 ***
StateWest Virginia 57.555148 8.042615 7.156 3.42e-12 ***
StateWisconsin 44.324909 4.925607 8.999 < 2e-16 ***
StateWyoming 49.925959 5.928977 8.421 5.14e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.74 on 447 degrees of freedom
Multiple R-squared: 0.9926, Adjusted R-squared: 0.9918
F-statistic: 1179 on 51 and 447 DF, p-value: < 2.2e-16Dmodel4 <- lm(nmfr ~ DDR + State -1, data=fouryeartestmerge)
summary(Dmodel4)
Call:
lm(formula = nmfr ~ DDR + State - 1, data = fouryeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.518 -1.709 -0.139 1.693 16.661
Coefficients:
Estimate Std. Error t value Pr(>|t|)
DDR -0.53571 0.04075 -13.14 <2e-16 ***
StateAlabama 73.36777 1.78884 41.01 <2e-16 ***
StateAlaska 85.23740 2.57022 33.16 <2e-16 ***
StateArizona 87.79768 2.25856 38.87 <2e-16 ***
StateArkansas 85.87312 1.83436 46.81 <2e-16 ***
StateCalifornia 65.29577 1.88319 34.67 <2e-16 ***
StateColorado 58.54537 2.16756 27.01 <2e-16 ***
StateConnecticut 49.64018 1.68151 29.52 <2e-16 ***
StateDelaware 67.15237 1.85619 36.18 <2e-16 ***
StateDistrict of Columbia 53.57838 1.75034 30.61 <2e-16 ***
StateFlorida 73.82707 1.92334 38.38 <2e-16 ***
StateGeorgia 73.29622 1.68444 43.51 <2e-16 ***
StateHawaii 65.55887 1.67408 39.16 <2e-16 ***
StateIdaho 67.67302 2.04297 33.12 <2e-16 ***
StateIllinois 63.92056 1.63274 39.15 <2e-16 ***
StateIndiana 76.36172 1.86898 40.86 <2e-16 ***
StateIowa 64.25450 1.67859 38.28 <2e-16 ***
StateKansas 73.03236 1.78798 40.85 <2e-16 ***
StateKentucky 86.31882 2.15373 40.08 <2e-16 ***
StateLouisiana 83.95261 1.83818 45.67 <2e-16 ***
StateMaine 60.09558 1.88365 31.90 <2e-16 ***
StateMaryland 58.27462 1.67593 34.77 <2e-16 ***
StateMassachusetts 55.34059 1.72647 32.05 <2e-16 ***
StateMichigan 65.94335 1.87646 35.14 <2e-16 ***
StateMinnesota 57.23052 1.66643 34.34 <2e-16 ***
StateMississippi 84.51259 1.73448 48.73 <2e-16 ***
StateMissouri 73.77827 1.90387 38.75 <2e-16 ***
StateMontana 73.49480 2.30014 31.95 <2e-16 ***
StateNebraska 63.40131 1.60902 39.40 <2e-16 ***
StateNevada 86.12854 2.29505 37.53 <2e-16 ***
StateNew Hampshire 53.95050 1.95903 27.54 <2e-16 ***
StateNew Jersey 50.25199 1.56713 32.07 <2e-16 ***
StateNew Mexico 95.11391 2.88761 32.94 <2e-16 ***
StateNew York 55.45688 1.56713 35.39 <2e-16 ***
StateNorth Carolina 68.78076 1.78569 38.52 <2e-16 ***
StateOhio 74.99415 1.92518 38.95 <2e-16 ***
StateOklahoma 95.67358 2.24444 42.63 <2e-16 ***
StateOregon 64.68885 2.12524 30.44 <2e-16 ***
StatePennsylvania 63.22829 1.86570 33.89 <2e-16 ***
StateRhode Island 57.90532 1.95748 29.58 <2e-16 ***
StateSouth Carolina 75.99210 1.84790 41.12 <2e-16 ***
StateSouth Dakota 77.28516 1.90994 40.47 <2e-16 ***
StateTennessee 80.65153 2.00904 40.14 <2e-16 ***
StateTexas 79.18766 1.66065 47.69 <2e-16 ***
StateUtah 60.62683 2.21100 27.42 <2e-16 ***
StateVermont 55.17017 1.89691 29.08 <2e-16 ***
StateVirginia 62.72147 1.63785 38.30 <2e-16 ***
StateWashington 63.50028 1.99677 31.80 <2e-16 ***
StateWest Virginia 86.45529 2.41088 35.86 <2e-16 ***
StateWisconsin 62.52909 1.81596 34.43 <2e-16 ***
StateWyoming 78.65679 2.43515 32.30 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.025 on 447 degrees of freedom
Multiple R-squared: 0.9947, Adjusted R-squared: 0.9941
F-statistic: 1638 on 51 and 447 DF, p-value: < 2.2e-16OPmodel4 <- lm(nmfr ~ OPDR + State -1, data=fouryeartestmerge)
summary(OPmodel4)
Call:
lm(formula = nmfr ~ OPDR + State - 1, data = fouryeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.868 -2.066 -0.238 1.869 17.368
Coefficients:
Estimate Std. Error t value Pr(>|t|)
OPDR -0.60723 0.07285 -8.336 9.58e-16 ***
StateAlabama 59.37266 1.42698 41.607 < 2e-16 ***
StateAlaska 61.36934 1.54179 39.804 < 2e-16 ***
StateArizona 68.58630 1.53320 44.734 < 2e-16 ***
StateArkansas 72.33642 1.46801 49.275 < 2e-16 ***
StateCalifornia 51.96071 1.59676 32.541 < 2e-16 ***
StateColorado 40.10423 1.50058 26.726 < 2e-16 ***
StateConnecticut 40.71117 1.54335 26.378 < 2e-16 ***
StateDelaware 55.58136 1.57970 35.185 < 2e-16 ***
StateDistrict of Columbia 42.38248 1.49951 28.264 < 2e-16 ***
StateFlorida 59.70733 1.51019 39.536 < 2e-16 ***
StateGeorgia 62.06155 1.44846 42.847 < 2e-16 ***
StateHawaii 54.03519 1.43345 37.696 < 2e-16 ***
StateIdaho 49.60697 1.43831 34.490 < 2e-16 ***
StateIllinois 55.16749 1.50356 36.691 < 2e-16 ***
StateIndiana 61.66388 1.44925 42.549 < 2e-16 ***
StateIowa 52.92515 1.44177 36.709 < 2e-16 ***
StateKansas 59.52697 1.44011 41.335 < 2e-16 ***
StateKentucky 71.60030 1.70088 42.096 < 2e-16 ***
StateLouisiana 69.07716 1.42775 48.382 < 2e-16 ***
StateMaine 47.72811 1.56291 30.538 < 2e-16 ***
StateMaryland 50.76365 1.61668 31.400 < 2e-16 ***
StateMassachusetts 47.33791 1.64849 28.716 < 2e-16 ***
StateMichigan 52.71466 1.51300 34.841 < 2e-16 ***
StateMinnesota 46.05044 1.43885 32.005 < 2e-16 ***
StateMississippi 71.17492 1.41807 50.191 < 2e-16 ***
StateMissouri 60.94865 1.55736 39.136 < 2e-16 ***
StateMontana 52.21574 1.47100 35.497 < 2e-16 ***
StateNebraska 52.12782 1.40864 37.006 < 2e-16 ***
StateNevada 69.90909 1.75506 39.833 < 2e-16 ***
StateNew Hampshire 42.43370 1.69672 25.009 < 2e-16 ***
StateNew Jersey 41.53942 1.44964 28.655 < 2e-16 ***
StateNew Mexico 69.81137 1.74669 39.968 < 2e-16 ***
StateNew York 47.62479 1.48201 32.135 < 2e-16 ***
StateNorth Carolina 57.78963 1.54117 37.497 < 2e-16 ***
StateOhio 63.10114 1.63351 38.629 < 2e-16 ***
StateOklahoma 79.30434 1.68805 46.980 < 2e-16 ***
StateOregon 47.72210 1.53778 31.033 < 2e-16 ***
StatePennsylvania 49.16625 1.46961 33.455 < 2e-16 ***
StateRhode Island 46.96182 1.73491 27.069 < 2e-16 ***
StateSouth Carolina 62.22113 1.46846 42.372 < 2e-16 ***
StateSouth Dakota 60.91757 1.42264 42.820 < 2e-16 ***
StateTennessee 66.56523 1.58868 41.900 < 2e-16 ***
StateTexas 67.82790 1.43047 47.417 < 2e-16 ***
StateUtah 45.93489 1.76886 25.969 < 2e-16 ***
StateVermont 42.34839 1.55124 27.300 < 2e-16 ***
StateVirginia 53.35119 1.48176 36.005 < 2e-16 ***
StateWashington 49.10069 1.55963 31.482 < 2e-16 ***
StateWest Virginia 74.48527 2.27120 32.796 < 2e-16 ***
StateWisconsin 50.56256 1.52074 33.249 < 2e-16 ***
StateWyoming 55.62459 1.48499 37.458 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.41 on 447 degrees of freedom
Multiple R-squared: 0.9936, Adjusted R-squared: 0.9929
F-statistic: 1363 on 51 and 447 DF, p-value: < 2.2e-1644. fixed and random effect models with four year lags and Hausman test
MLfixed4 <- plm(nmfr ~ MLDR, data=fouryeartestmerge, index=c("State", "year"), model="within")
MLrandom4 <- plm(nmfr ~ MLDR, data=fouryeartestmerge, index=c("State", "year"), model="random")
summary(MLfixed4)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ MLDR, data = fouryeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-10, N = 498
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.914380 -2.512200 -0.037569 2.378607 17.188634
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
MLDR 0.0041219 0.0163977 0.2514 0.8016
Total Sum of Squares: 10043
Residual Sum of Squares: 10041
R-Squared: 0.00014134
Adj. R-Squared: -0.1117
F-statistic: 0.0631883 on 1 and 447 DF, p-value: 0.80164summary(MLrandom4)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ MLDR, data = fouryeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-10, N = 498
Effects:
var std.dev share
idiosyncratic 22.46 4.74 0.332
individual 45.29 6.73 0.668
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.7584 0.7826 0.7826 0.7822 0.7826 0.7826
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-47.666 -2.836 -0.105 -0.001 2.653 19.700
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 29.937223 3.908026 7.6604 1.853e-14 ***
MLDR 0.060929 0.010865 5.6078 2.050e-08 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 12321
Residual Sum of Squares: 11584
R-Squared: 0.059797
Adj. R-Squared: 0.057901
Chisq: 31.447 on 1 DF, p-value: 2.0496e-08phtest(MLfixed4, MLrandom4)
Hausman Test
data: nmfr ~ MLDR
chisq = 21.395, df = 1, p-value = 3.738e-06
alternative hypothesis: one model is inconsistentDfixed4 <- plm(nmfr ~ DDR, data=fouryeartestmerge, index=c("State", "year"), model="within")
Drandom4 <- plm(nmfr ~ DDR, data=fouryeartestmerge, index=c("State", "year"), model="random")
summary(Dfixed4)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ DDR, data = fouryeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-10, N = 498
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.51797 -1.70916 -0.13883 1.69271 16.66118
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
DDR -0.535706 0.040753 -13.145 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 10043
Residual Sum of Squares: 7242.9
R-Squared: 0.27879
Adj. R-Squared: 0.19812
F-statistic: 172.795 on 1 and 447 DF, p-value: < 2.22e-16summary(Drandom4)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ DDR, data = fouryeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-10, N = 498
Effects:
var std.dev share
idiosyncratic 16.203 4.025 0.147
individual 94.075 9.699 0.853
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.8548 0.8699 0.8699 0.8696 0.8699 0.8699
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-49.195 -2.219 -0.344 0.001 1.819 17.977
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 68.19447 1.99969 34.103 < 2.2e-16 ***
DDR -0.48665 0.04033 -12.067 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 10859
Residual Sum of Squares: 8394
R-Squared: 0.22702
Adj. R-Squared: 0.22546
Chisq: 145.609 on 1 DF, p-value: < 2.22e-16phtest(Dfixed4, Drandom4)
Hausman Test
data: nmfr ~ DDR
chisq = 70.057, df = 1, p-value < 2.2e-16
alternative hypothesis: one model is inconsistentOPfixed4 <- plm(nmfr ~ OPDR, data=fouryeartestmerge, index=c("State", "year"), model="within")
OPrandom4 <- plm(nmfr ~ OPDR, data=fouryeartestmerge, index=c("State", "year"), model="random")
summary(OPfixed4)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ OPDR, data = fouryeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-10, N = 498
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.86795 -2.06561 -0.23769 1.86900 17.36798
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
OPDR -0.607227 0.072846 -8.3358 9.581e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 10043
Residual Sum of Squares: 8691.6
R-Squared: 0.13453
Adj. R-Squared: 0.037727
F-statistic: 69.4852 on 1 and 447 DF, p-value: 9.5808e-16summary(OPrandom4)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ OPDR, data = fouryeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-10, N = 498
Effects:
var std.dev share
idiosyncratic 19.444 4.410 0.165
individual 98.318 9.916 0.835
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.8447 0.8607 0.8607 0.8605 0.8607 0.8607
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-49.529 -2.293 -0.308 0.001 2.032 18.985
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 56.113902 1.539727 36.4441 < 2.2e-16 ***
OPDR -0.589556 0.071396 -8.2576 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 10978
Residual Sum of Squares: 9650.1
R-Squared: 0.12095
Adj. R-Squared: 0.11918
Chisq: 68.1882 on 1 DF, p-value: < 2.22e-16phtest(OPfixed4, OPrandom4)
Hausman Test
data: nmfr ~ OPDR
chisq = 1.4926, df = 1, p-value = 0.2218
alternative hypothesis: one model is inconsistent45. fixed effect models with states included with three year lags
MLmodel3 <- lm(nmfr ~ MLDR + State -1, data=threeyeartestmerge)
summary(MLmodel3)
Call:
lm(formula = nmfr ~ MLDR + State - 1, data = threeyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-49.254 -2.191 -0.169 2.128 16.374
Coefficients:
Estimate Std. Error t value Pr(>|t|)
MLDR -0.03190 0.01843 -1.730 0.0844 .
StateAlabama 71.70224 8.95049 8.011 1.27e-14 ***
StateAlaska 67.19365 6.55143 10.256 < 2e-16 ***
StateArizona 75.15194 6.40554 11.732 < 2e-16 ***
StateArkansas 83.98529 8.70096 9.652 < 2e-16 ***
StateCalifornia 57.82413 5.30550 10.899 < 2e-16 ***
StateColorado 44.79577 5.39026 8.310 1.52e-15 ***
StateConnecticut 43.86947 5.03529 8.712 < 2e-16 ***
StateDelaware 61.23843 6.81103 8.991 < 2e-16 ***
StateDistrict of Columbia 52.99859 8.57215 6.183 1.56e-09 ***
StateFlorida 66.70330 6.64209 10.043 < 2e-16 ***
StateGeorgia 71.50239 7.12835 10.031 < 2e-16 ***
StateHawaii 60.90255 5.41698 11.243 < 2e-16 ***
StateIdaho 56.54526 5.71158 9.900 < 2e-16 ***
StateIllinois 61.34318 5.96558 10.283 < 2e-16 ***
StateIndiana 71.00012 7.06515 10.049 < 2e-16 ***
StateIowa 59.82338 5.69027 10.513 < 2e-16 ***
StateKansas 67.82926 6.30296 10.761 < 2e-16 ***
StateKentucky 78.43464 8.48873 9.240 < 2e-16 ***
StateLouisiana 81.97043 8.68705 9.436 < 2e-16 ***
StateMaine 52.32121 5.82335 8.985 < 2e-16 ***
StateMaryland 54.42860 6.11795 8.897 < 2e-16 ***
StateMassachusetts 48.95533 5.14625 9.513 < 2e-16 ***
StateMichigan 59.71917 6.76453 8.828 < 2e-16 ***
StateMinnesota 51.41812 4.78074 10.755 < 2e-16 ***
StateMississippi 85.64897 9.49646 9.019 < 2e-16 ***
StateMissouri 67.85859 7.25183 9.357 < 2e-16 ***
StateMontana 60.18766 6.51041 9.245 < 2e-16 ***
StateNebraska 60.20992 5.54600 10.856 < 2e-16 ***
StateNevada 73.14422 6.94182 10.537 < 2e-16 ***
StateNew Hampshire 43.48029 5.20907 8.347 1.17e-15 ***
StateNew Jersey 47.42426 5.33608 8.887 < 2e-16 ***
StateNew Mexico 74.69745 7.48625 9.978 < 2e-16 ***
StateNew York 52.72729 5.31940 9.912 < 2e-16 ***
StateNorth Carolina 64.54916 6.88210 9.379 < 2e-16 ***
StateOhio 68.23749 7.06095 9.664 < 2e-16 ***
StateOklahoma 87.26194 8.75702 9.965 < 2e-16 ***
StateOregon 53.04865 5.90462 8.984 < 2e-16 ***
StatePennsylvania 56.43706 6.41290 8.801 < 2e-16 ***
StateRhode Island 48.30325 5.70388 8.468 4.85e-16 ***
StateSouth Carolina 72.55703 7.99764 9.072 < 2e-16 ***
StateSouth Dakota 68.75584 5.89176 11.670 < 2e-16 ***
StateTennessee 74.92001 8.34874 8.974 < 2e-16 ***
StateTexas 77.22170 6.44752 11.977 < 2e-16 ***
StateUtah 46.78073 5.51589 8.481 4.42e-16 ***
StateVermont 45.71110 5.23746 8.728 < 2e-16 ***
StateVirginia 59.39787 5.84352 10.165 < 2e-16 ***
StateWashington 52.92652 5.40225 9.797 < 2e-16 ***
StateWest Virginia 75.05049 9.06817 8.276 1.94e-15 ***
StateWisconsin 55.10783 5.49464 10.029 < 2e-16 ***
StateWyoming 62.93449 6.62355 9.502 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.59 on 398 degrees of freedom
Multiple R-squared: 0.9933, Adjusted R-squared: 0.9924
F-statistic: 1156 on 51 and 398 DF, p-value: < 2.2e-16Dmodel3 <- lm(nmfr ~ DDR + State -1, data=threeyeartestmerge)
summary(Dmodel3)
Call:
lm(formula = nmfr ~ DDR + State - 1, data = threeyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.416 -1.580 -0.170 1.329 15.213
Coefficients:
Estimate Std. Error t value Pr(>|t|)
DDR -0.50662 0.04334 -11.69 <2e-16 ***
StateAlabama 72.34527 1.89910 38.09 <2e-16 ***
StateAlaska 84.09702 2.73194 30.78 <2e-16 ***
StateArizona 87.87937 2.40713 36.51 <2e-16 ***
StateArkansas 85.81539 1.94549 44.11 <2e-16 ***
StateCalifornia 64.59280 1.93278 33.42 <2e-16 ***
StateColorado 58.01428 2.31309 25.08 <2e-16 ***
StateConnecticut 49.42763 1.77806 27.80 <2e-16 ***
StateDelaware 66.99072 1.98246 33.79 <2e-16 ***
StateDistrict of Columbia 52.94426 1.81743 29.13 <2e-16 ***
StateFlorida 73.52134 2.03165 36.19 <2e-16 ***
StateGeorgia 73.32006 1.77838 41.23 <2e-16 ***
StateHawaii 65.75871 1.77742 37.00 <2e-16 ***
StateIdaho 67.29400 2.18243 30.83 <2e-16 ***
StateIllinois 64.22300 1.72208 37.29 <2e-16 ***
StateIndiana 76.38604 1.98784 38.43 <2e-16 ***
StateIowa 64.21576 1.77902 36.10 <2e-16 ***
StateKansas 72.99396 1.88946 38.63 <2e-16 ***
StateKentucky 85.97994 2.30127 37.36 <2e-16 ***
StateLouisiana 83.65611 1.93494 43.23 <2e-16 ***
StateMaine 60.15051 2.00260 30.04 <2e-16 ***
StateMaryland 57.77892 1.76336 32.77 <2e-16 ***
StateMassachusetts 55.12841 1.82536 30.20 <2e-16 ***
StateMichigan 65.70603 1.99215 32.98 <2e-16 ***
StateMinnesota 57.19779 1.76432 32.42 <2e-16 ***
StateMississippi 84.12378 1.82636 46.06 <2e-16 ***
StateMissouri 73.41484 2.01999 36.34 <2e-16 ***
StateMontana 73.34751 2.45153 29.92 <2e-16 ***
StateNebraska 63.39254 1.69774 37.34 <2e-16 ***
StateNevada 85.43616 2.44424 34.95 <2e-16 ***
StateNew Hampshire 53.88525 2.09849 25.68 <2e-16 ***
StateNew Jersey 50.05096 1.64894 30.35 <2e-16 ***
StateNew Mexico 94.76736 3.09932 30.58 <2e-16 ***
StateNew York 55.41789 1.65066 33.57 <2e-16 ***
StateNorth Carolina 68.60477 1.88466 36.40 <2e-16 ***
StateOhio 74.63846 2.05290 36.36 <2e-16 ***
StateOklahoma 95.70338 2.39789 39.91 <2e-16 ***
StateOregon 64.55444 2.25816 28.59 <2e-16 ***
StatePennsylvania 62.99316 1.98856 31.68 <2e-16 ***
StateRhode Island 57.55159 2.08099 27.66 <2e-16 ***
StateSouth Carolina 75.74892 1.95326 38.78 <2e-16 ***
StateSouth Dakota 76.79463 2.02399 37.94 <2e-16 ***
StateTennessee 80.17290 2.12737 37.69 <2e-16 ***
StateTexas 79.70260 1.74757 45.61 <2e-16 ***
StateUtah 60.34694 2.35352 25.64 <2e-16 ***
StateVermont 54.73057 2.01128 27.21 <2e-16 ***
StateVirginia 62.63146 1.72949 36.21 <2e-16 ***
StateWashington 63.30324 2.11985 29.86 <2e-16 ***
StateWest Virginia 85.63051 2.59245 33.03 <2e-16 ***
StateWisconsin 62.29516 1.92479 32.37 <2e-16 ***
StateWyoming 77.95891 2.60198 29.96 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 3.975 on 398 degrees of freedom
Multiple R-squared: 0.995, Adjusted R-squared: 0.9943
F-statistic: 1544 on 51 and 398 DF, p-value: < 2.2e-16OPmodel3 <- lm(nmfr ~ OPDR + State -1, data=threeyeartestmerge)
summary(OPmodel3)
Call:
lm(formula = nmfr ~ OPDR + State - 1, data = threeyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.884 -1.824 -0.302 1.709 15.988
Coefficients:
Estimate Std. Error t value Pr(>|t|)
OPDR -0.55151 0.07562 -7.293 1.65e-12 ***
StateAlabama 58.82685 1.47894 39.776 < 2e-16 ***
StateAlaska 61.45335 1.61399 38.075 < 2e-16 ***
StateArizona 69.27311 1.59024 43.562 < 2e-16 ***
StateArkansas 72.71259 1.52220 47.768 < 2e-16 ***
StateCalifornia 51.75878 1.57322 32.900 < 2e-16 ***
StateColorado 40.10531 1.55578 25.778 < 2e-16 ***
StateConnecticut 40.74074 1.60726 25.348 < 2e-16 ***
StateDelaware 55.72301 1.65835 33.601 < 2e-16 ***
StateDistrict of Columbia 42.45485 1.54560 27.468 < 2e-16 ***
StateFlorida 59.97449 1.56638 38.289 < 2e-16 ***
StateGeorgia 62.53182 1.50278 41.611 < 2e-16 ***
StateHawaii 54.46035 1.48412 36.695 < 2e-16 ***
StateIdaho 49.68317 1.48776 33.395 < 2e-16 ***
StateIllinois 55.67406 1.55863 35.720 < 2e-16 ***
StateIndiana 62.21507 1.50564 41.321 < 2e-16 ***
StateIowa 53.22017 1.49546 35.588 < 2e-16 ***
StateKansas 60.00082 1.49089 40.245 < 2e-16 ***
StateKentucky 71.68983 1.78769 40.102 < 2e-16 ***
StateLouisiana 69.47210 1.47639 47.055 < 2e-16 ***
StateMaine 48.08959 1.62701 29.557 < 2e-16 ***
StateMaryland 50.46624 1.68034 30.033 < 2e-16 ***
StateMassachusetts 47.23080 1.71579 27.527 < 2e-16 ***
StateMichigan 52.88917 1.57267 33.630 < 2e-16 ***
StateMinnesota 46.38147 1.49261 31.074 < 2e-16 ***
StateMississippi 71.46019 1.46894 48.647 < 2e-16 ***
StateMissouri 60.99180 1.62122 37.621 < 2e-16 ***
StateMontana 52.85972 1.52546 34.652 < 2e-16 ***
StateNebraska 52.54865 1.45820 36.037 < 2e-16 ***
StateNevada 69.57205 1.82408 38.141 < 2e-16 ***
StateNew Hampshire 42.51215 1.78323 23.840 < 2e-16 ***
StateNew Jersey 41.64894 1.50254 27.719 < 2e-16 ***
StateNew Mexico 70.10001 1.81894 38.539 < 2e-16 ***
StateNew York 47.81549 1.53847 31.080 < 2e-16 ***
StateNorth Carolina 57.98226 1.59956 36.249 < 2e-16 ***
StateOhio 63.12435 1.71853 36.732 < 2e-16 ***
StateOklahoma 79.66548 1.75741 45.331 < 2e-16 ***
StateOregon 48.04633 1.58918 30.233 < 2e-16 ***
StatePennsylvania 49.35019 1.52849 32.287 < 2e-16 ***
StateRhode Island 46.73106 1.80673 25.865 < 2e-16 ***
StateSouth Carolina 62.54680 1.52328 41.061 < 2e-16 ***
StateSouth Dakota 61.08030 1.47272 41.474 < 2e-16 ***
StateTennessee 66.65056 1.65587 40.251 < 2e-16 ***
StateTexas 68.78631 1.47950 46.493 < 2e-16 ***
StateUtah 45.91432 1.83753 24.987 < 2e-16 ***
StateVermont 42.18527 1.60541 26.277 < 2e-16 ***
StateVirginia 53.53073 1.53789 34.808 < 2e-16 ***
StateWashington 49.22554 1.61286 30.521 < 2e-16 ***
StateWest Virginia 73.67881 2.41137 30.555 < 2e-16 ***
StateWisconsin 50.71328 1.58185 32.059 < 2e-16 ***
StateWyoming 55.91386 1.54956 36.084 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.327 on 398 degrees of freedom
Multiple R-squared: 0.994, Adjusted R-squared: 0.9933
F-statistic: 1302 on 51 and 398 DF, p-value: < 2.2e-1646. fixed and random effect models with three year lags and Hausman test
MLfixed3 <- plm(nmfr ~ MLDR, data=threeyeartestmerge, index=c("State", "year"), model="within")
MLrandom3 <- plm(nmfr ~ MLDR, data=threeyeartestmerge, index=c("State", "year"), model="random")
summary(MLfixed3)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ MLDR, data = threeyeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-9, N = 449
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-49.25410 -2.19056 -0.16943 2.12793 16.37356
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
MLDR -0.031897 0.018434 -1.7303 0.08435 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 8448.2
Residual Sum of Squares: 8385.1
R-Squared: 0.0074663
Adj. R-Squared: -0.11722
F-statistic: 2.99395 on 1 and 398 DF, p-value: 0.084351summary(MLrandom3)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ MLDR, data = threeyeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-9, N = 449
Effects:
var std.dev share
idiosyncratic 21.068 4.590 0.316
individual 45.534 6.748 0.684
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.7662 0.7789 0.7789 0.7786 0.7789 0.7789
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-48.17 -2.64 -0.26 0.00 2.35 18.76
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 33.083769 4.148339 7.9752 1.522e-15 ***
MLDR 0.053539 0.011594 4.6178 3.879e-06 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 10591
Residual Sum of Squares: 10112
R-Squared: 0.045268
Adj. R-Squared: 0.043132
Chisq: 21.3239 on 1 DF, p-value: 3.8786e-06phtest(MLfixed3, MLrandom3)
Hausman Test
data: nmfr ~ MLDR
chisq = 35.536, df = 1, p-value = 2.503e-09
alternative hypothesis: one model is inconsistentDfixed3 <- plm(nmfr ~ DDR, data=threeyeartestmerge, index=c("State", "year"), model="within")
Drandom3 <- plm(nmfr ~ DDR, data=threeyeartestmerge, index=c("State", "year"), model="random")
summary(Dfixed3)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ DDR, data = threeyeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-9, N = 449
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.41568 -1.57989 -0.17032 1.32913 15.21348
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
DDR -0.506617 0.043342 -11.689 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 8448.2
Residual Sum of Squares: 6289.2
R-Squared: 0.25556
Adj. R-Squared: 0.16204
F-statistic: 136.632 on 1 and 398 DF, p-value: < 2.22e-16summary(Drandom3)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ DDR, data = threeyeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-9, N = 449
Effects:
var std.dev share
idiosyncratic 15.802 3.975 0.143
individual 94.954 9.744 0.857
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.8572 0.8653 0.8653 0.8651 0.8653 0.8653
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-49.159 -2.110 -0.509 0.001 1.528 16.721
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 67.731188 2.079657 32.568 < 2.2e-16 ***
DDR -0.452522 0.042712 -10.595 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 9244
Residual Sum of Squares: 7389.5
R-Squared: 0.20062
Adj. R-Squared: 0.19883
Chisq: 112.247 on 1 DF, p-value: < 2.22e-16phtest(Dfixed3, Drandom3)
Hausman Test
data: nmfr ~ DDR
chisq = 54.037, df = 1, p-value = 1.968e-13
alternative hypothesis: one model is inconsistentOPfixed3 <- plm(nmfr ~ OPDR, data=threeyeartestmerge, index=c("State", "year"), model="within")
OPrandom3 <- plm(nmfr ~ OPDR, data=threeyeartestmerge, index=c("State", "year"), model="random")
summary(OPfixed3)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ OPDR, data = threeyeartestmerge, model = "within",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-9, N = 449
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.88390 -1.82448 -0.30185 1.70882 15.98780
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
OPDR -0.551507 0.075616 -7.2935 1.646e-12 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 8448.2
Residual Sum of Squares: 7452.2
R-Squared: 0.1179
Adj. R-Squared: 0.0070815
F-statistic: 53.1951 on 1 and 398 DF, p-value: 1.6459e-12summary(OPrandom3)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ OPDR, data = threeyeartestmerge, model = "random",
index = c("State", "year"))
Unbalanced Panel: n = 50, T = 8-9, N = 449
Effects:
var std.dev share
idiosyncratic 18.724 4.327 0.159
individual 99.345 9.967 0.841
theta:
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.8483 0.8568 0.8568 0.8566 0.8568 0.8568
Residuals:
Min. 1st Qu. Median Mean 3rd Qu. Max.
-49.585 -2.181 -0.499 0.001 1.836 17.769
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 56.27540 1.56173 36.0340 < 2.2e-16 ***
OPDR -0.53325 0.07391 -7.2148 5.401e-13 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 9347.3
Residual Sum of Squares: 8373.4
R-Squared: 0.10419
Adj. R-Squared: 0.10218
Chisq: 52.0533 on 1 DF, p-value: 5.4014e-13phtest(OPfixed3, OPrandom3)
Hausman Test
data: nmfr ~ OPDR
chisq = 1.307, df = 1, p-value = 0.2529
alternative hypothesis: one model is inconsistent47. fixed effect models with states included with two year lags
MLmodel2 <- lm(nmfr ~ MLDR + State -1, data=twoyeartestmerge)
summary(MLmodel2)
Call:
lm(formula = nmfr ~ MLDR + State - 1, data = twoyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-49.149 -1.895 -0.331 1.662 15.232
Coefficients:
Estimate Std. Error t value Pr(>|t|)
MLDR -0.06063 0.01970 -3.078 0.00225 **
StateAlabama 85.41663 9.56219 8.933 < 2e-16 ***
StateAlaska 77.93943 6.98305 11.161 < 2e-16 ***
StateArizona 85.75406 6.80738 12.597 < 2e-16 ***
StateArkansas 98.52175 9.29987 10.594 < 2e-16 ***
StateCalifornia 65.50307 5.55125 11.800 < 2e-16 ***
StateColorado 53.41756 5.72961 9.323 < 2e-16 ***
StateConnecticut 51.56976 5.34012 9.657 < 2e-16 ***
StateDelaware 71.92989 7.25948 9.908 < 2e-16 ***
StateDistrict of Columbia 66.38210 8.95309 7.414 9.34e-13 ***
StateFlorida 77.38273 7.04166 10.989 < 2e-16 ***
StateGeorgia 83.10657 7.56992 10.979 < 2e-16 ***
StateHawaii 69.61216 5.76210 12.081 < 2e-16 ***
StateIdaho 66.25850 6.10806 10.848 < 2e-16 ***
StateIllinois 70.95797 6.32441 11.220 < 2e-16 ***
StateIndiana 82.35330 7.56630 10.884 < 2e-16 ***
StateIowa 68.91410 6.07090 11.352 < 2e-16 ***
StateKansas 78.06886 6.72514 11.609 < 2e-16 ***
StateKentucky 92.21942 9.11272 10.120 < 2e-16 ***
StateLouisiana 96.04384 9.23092 10.405 < 2e-16 ***
StateMaine 61.32946 6.21081 9.875 < 2e-16 ***
StateMaryland 63.84398 6.45774 9.886 < 2e-16 ***
StateMassachusetts 56.99461 5.45103 10.456 < 2e-16 ***
StateMichigan 70.54678 7.20896 9.786 < 2e-16 ***
StateMinnesota 58.87329 5.08444 11.579 < 2e-16 ***
StateMississippi 100.75007 10.13705 9.939 < 2e-16 ***
StateMissouri 79.20370 7.76528 10.200 < 2e-16 ***
StateMontana 70.72769 6.95880 10.164 < 2e-16 ***
StateNebraska 69.08573 5.90792 11.694 < 2e-16 ***
StateNevada 83.86752 7.36273 11.391 < 2e-16 ***
StateNew Hampshire 51.62781 5.54936 9.303 < 2e-16 ***
StateNew Jersey 55.62928 5.65139 9.843 < 2e-16 ***
StateNew Mexico 87.37512 8.00892 10.910 < 2e-16 ***
StateNew York 61.01103 5.61752 10.861 < 2e-16 ***
StateNorth Carolina 75.63109 7.30207 10.357 < 2e-16 ***
StateOhio 79.63190 7.55112 10.546 < 2e-16 ***
StateOklahoma 101.91837 9.37587 10.870 < 2e-16 ***
StateOregon 62.61335 6.26759 9.990 < 2e-16 ***
StatePennsylvania 66.44041 6.82632 9.733 < 2e-16 ***
StateRhode Island 57.42058 6.06566 9.467 < 2e-16 ***
StateSouth Carolina 85.20383 8.49436 10.031 < 2e-16 ***
StateSouth Dakota 78.87424 6.31008 12.500 < 2e-16 ***
StateTennessee 88.38217 8.91914 9.909 < 2e-16 ***
StateTexas 88.34205 6.83976 12.916 < 2e-16 ***
StateUtah 55.72185 5.84400 9.535 < 2e-16 ***
StateVermont 54.18258 5.56710 9.733 < 2e-16 ***
StateVirginia 68.39231 6.19149 11.046 < 2e-16 ***
StateWashington 61.59014 5.73482 10.740 < 2e-16 ***
StateWest Virginia 89.75427 9.70145 9.252 < 2e-16 ***
StateWisconsin 63.72174 5.83593 10.919 < 2e-16 ***
StateWyoming 74.01335 7.03565 10.520 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.352 on 349 degrees of freedom
Multiple R-squared: 0.9942, Adjusted R-squared: 0.9934
F-statistic: 1173 on 51 and 349 DF, p-value: < 2.2e-16Dmodel2 <- lm(nmfr ~ DDR + State -1, data=twoyeartestmerge)
summary(Dmodel2)
Call:
lm(formula = nmfr ~ DDR + State - 1, data = twoyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.103 -1.555 -0.286 1.432 14.069
Coefficients:
Estimate Std. Error t value Pr(>|t|)
DDR -0.4488 0.0460 -9.756 <2e-16 ***
StateAlabama 70.6469 2.0133 35.091 <2e-16 ***
StateAlaska 81.8729 2.9025 28.208 <2e-16 ***
StateArizona 86.4165 2.5645 33.697 <2e-16 ***
StateArkansas 85.2380 2.0636 41.307 <2e-16 ***
StateCalifornia 62.9847 1.9850 31.730 <2e-16 ***
StateColorado 56.2411 2.4565 22.895 <2e-16 ***
StateConnecticut 48.2984 1.8825 25.656 <2e-16 ***
StateDelaware 65.7864 2.1210 31.017 <2e-16 ***
StateDistrict of Columbia 52.2909 1.9240 27.178 <2e-16 ***
StateFlorida 72.1895 2.1415 33.709 <2e-16 ***
StateGeorgia 72.6877 1.8775 38.716 <2e-16 ***
StateHawaii 64.9383 1.8790 34.559 <2e-16 ***
StateIdaho 66.4562 2.3377 28.428 <2e-16 ***
StateIllinois 63.6638 1.8225 34.933 <2e-16 ***
StateIndiana 75.2386 2.1205 35.481 <2e-16 ***
StateIowa 63.4241 1.8908 33.544 <2e-16 ***
StateKansas 71.9904 1.9978 36.034 <2e-16 ***
StateKentucky 84.4453 2.4617 34.303 <2e-16 ***
StateLouisiana 82.5154 2.0288 40.672 <2e-16 ***
StateMaine 58.4967 2.1205 27.586 <2e-16 ***
StateMaryland 56.5757 1.8528 30.536 <2e-16 ***
StateMassachusetts 54.0178 1.9284 28.011 <2e-16 ***
StateMichigan 64.5568 2.1205 30.444 <2e-16 ***
StateMinnesota 56.2743 1.8720 30.060 <2e-16 ***
StateMississippi 82.9153 1.9204 43.175 <2e-16 ***
StateMissouri 71.8957 2.1543 33.373 <2e-16 ***
StateMontana 71.4953 2.6151 27.340 <2e-16 ***
StateNebraska 62.7741 1.7998 34.879 <2e-16 ***
StateNevada 83.2105 2.6019 31.981 <2e-16 ***
StateNew Hampshire 52.4364 2.2419 23.389 <2e-16 ***
StateNew Jersey 49.2731 1.7444 28.246 <2e-16 ***
StateNew Mexico 92.4459 3.3028 27.991 <2e-16 ***
StateNew York 54.7071 1.7409 31.425 <2e-16 ***
StateNorth Carolina 67.6649 1.9929 33.954 <2e-16 ***
StateOhio 73.4625 2.1911 33.528 <2e-16 ***
StateOklahoma 94.3549 2.5500 37.002 <2e-16 ***
StateOregon 62.9716 2.3932 26.313 <2e-16 ***
StatePennsylvania 61.6913 2.1231 29.057 <2e-16 ***
StateRhode Island 56.4911 2.2347 25.280 <2e-16 ***
StateSouth Carolina 74.5331 2.0708 35.992 <2e-16 ***
StateSouth Dakota 76.4106 2.1742 35.144 <2e-16 ***
StateTennessee 78.7726 2.2591 34.868 <2e-16 ***
StateTexas 79.7135 1.8425 43.265 <2e-16 ***
StateUtah 58.7074 2.5014 23.470 <2e-16 ***
StateVermont 53.6553 2.1402 25.070 <2e-16 ***
StateVirginia 61.5596 1.8251 33.730 <2e-16 ***
StateWashington 61.8390 2.2446 27.550 <2e-16 ***
StateWest Virginia 83.7164 2.7725 30.196 <2e-16 ***
StateWisconsin 60.9719 2.0355 29.954 <2e-16 ***
StateWyoming 76.4569 2.7839 27.464 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 3.909 on 349 degrees of freedom
Multiple R-squared: 0.9953, Adjusted R-squared: 0.9946
F-statistic: 1455 on 51 and 349 DF, p-value: < 2.2e-16OPmodel2 <- lm(nmfr ~ OPDR + State -1, data=twoyeartestmerge)
summary(OPmodel2)
Call:
lm(formula = nmfr ~ OPDR + State - 1, data = twoyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-48.967 -1.685 -0.291 1.469 14.370
Coefficients:
Estimate Std. Error t value Pr(>|t|)
OPDR -0.47350 0.07831 -6.046 3.82e-09 ***
StateAlabama 58.45401 1.52320 38.376 < 2e-16 ***
StateAlaska 61.94334 1.69615 36.520 < 2e-16 ***
StateArizona 69.60248 1.64241 42.378 < 2e-16 ***
StateArkansas 73.39564 1.56999 46.749 < 2e-16 ***
StateCalifornia 51.40410 1.53232 33.547 < 2e-16 ***
StateColorado 40.12106 1.60433 25.008 < 2e-16 ***
StateConnecticut 40.40925 1.66578 24.258 < 2e-16 ***
StateDelaware 55.55351 1.73700 31.983 < 2e-16 ***
StateDistrict of Columbia 42.83615 1.59880 26.793 < 2e-16 ***
StateFlorida 60.04757 1.61227 37.244 < 2e-16 ***
StateGeorgia 63.01887 1.55061 40.641 < 2e-16 ***
StateHawaii 54.64338 1.52455 35.842 < 2e-16 ***
StateIdaho 50.37642 1.53207 32.881 < 2e-16 ***
StateIllinois 55.90962 1.61342 34.653 < 2e-16 ***
StateIndiana 62.35983 1.55466 40.112 < 2e-16 ***
StateIowa 53.40475 1.54313 34.608 < 2e-16 ***
StateKansas 60.31284 1.53555 39.278 < 2e-16 ***
StateKentucky 71.53069 1.87778 38.093 < 2e-16 ***
StateLouisiana 69.95514 1.51743 46.101 < 2e-16 ***
StateMaine 47.63027 1.68442 28.277 < 2e-16 ***
StateMaryland 50.00037 1.73700 28.786 < 2e-16 ***
StateMassachusetts 46.85697 1.78116 26.307 < 2e-16 ***
StateMichigan 52.98008 1.63206 32.462 < 2e-16 ***
StateMinnesota 46.45912 1.54020 30.164 < 2e-16 ***
StateMississippi 71.67033 1.51190 47.404 < 2e-16 ***
StateMissouri 60.59142 1.68396 35.982 < 2e-16 ***
StateMontana 52.96151 1.57095 33.713 < 2e-16 ***
StateNebraska 52.89402 1.50062 35.248 < 2e-16 ***
StateNevada 68.72855 1.88440 36.472 < 2e-16 ***
StateNew Hampshire 41.95616 1.85580 22.608 < 2e-16 ***
StateNew Jersey 41.62404 1.55090 26.839 < 2e-16 ***
StateNew Mexico 70.18808 1.88259 37.283 < 2e-16 ***
StateNew York 47.84217 1.59019 30.086 < 2e-16 ***
StateNorth Carolina 58.04336 1.65134 35.149 < 2e-16 ***
StateOhio 63.08942 1.80479 34.957 < 2e-16 ***
StateOklahoma 79.69787 1.80814 44.077 < 2e-16 ***
StateOregon 48.05313 1.63411 29.406 < 2e-16 ***
StatePennsylvania 49.29380 1.58222 31.155 < 2e-16 ***
StateRhode Island 46.55040 1.90143 24.482 < 2e-16 ***
StateSouth Carolina 62.67228 1.57420 39.812 < 2e-16 ***
StateSouth Dakota 61.98329 1.51805 40.831 < 2e-16 ***
StateTennessee 66.60547 1.72041 38.715 < 2e-16 ***
StateTexas 69.89233 1.52232 45.912 < 2e-16 ***
StateUtah 45.51657 1.89715 23.992 < 2e-16 ***
StateVermont 42.15764 1.65566 25.463 < 2e-16 ***
StateVirginia 53.31035 1.58530 33.628 < 2e-16 ***
StateWashington 49.09374 1.66047 29.566 < 2e-16 ***
StateWest Virginia 72.75654 2.54308 28.610 < 2e-16 ***
StateWisconsin 50.57104 1.63658 30.900 < 2e-16 ***
StateWyoming 56.59652 1.60583 35.244 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 4.196 on 349 degrees of freedom
Multiple R-squared: 0.9946, Adjusted R-squared: 0.9938
F-statistic: 1262 on 51 and 349 DF, p-value: < 2.2e-1648. fixed and random effect models with two year lags and Hausman test
MLfixed2 <- plm(nmfr ~ MLDR, data=twoyeartestmerge, index=c("State", "year"), model="within")
MLrandom2 <- plm(nmfr ~ MLDR, data=twoyeartestmerge, index=c("State", "year"), model="random")
summary(MLfixed2)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ MLDR, data = twoyeartestmerge, model = "within",
index = c("State", "year"))
Balanced Panel: n = 50, T = 8, N = 400
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-49.14927 -1.89474 -0.33096 1.66208 15.23224
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
MLDR -0.060629 0.019696 -3.0783 0.002247 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 6788
Residual Sum of Squares: 6608.6
R-Squared: 0.026434
Adj. R-Squared: -0.11305
F-statistic: 9.476 on 1 and 349 DF, p-value: 0.0022466summary(MLrandom2)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ MLDR, data = twoyeartestmerge, model = "random",
index = c("State", "year"))
Balanced Panel: n = 50, T = 8, N = 400
Effects:
var std.dev share
idiosyncratic 18.936 4.352 0.291
individual 46.148 6.793 0.709
theta: 0.7791
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.7064 -2.4366 -0.2995 2.3147 17.5751
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 35.378015 4.330693 8.1691 3.106e-16 ***
MLDR 0.048713 0.012128 4.0168 5.900e-05 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 8742.5
Residual Sum of Squares: 8401.9
R-Squared: 0.038959
Adj. R-Squared: 0.036545
Chisq: 16.1343 on 1 DF, p-value: 5.9004e-05phtest(MLfixed2, MLrandom2)
Hausman Test
data: nmfr ~ MLDR
chisq = 49.642, df = 1, p-value = 1.845e-12
alternative hypothesis: one model is inconsistentDfixed2 <- plm(nmfr ~ DDR, data=twoyeartestmerge, index=c("State", "year"), model="within")
Drandom2 <- plm(nmfr ~ DDR, data=twoyeartestmerge, index=c("State", "year"), model="random")
summary(Dfixed2)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ DDR, data = twoyeartestmerge, model = "within",
index = c("State", "year"))
Balanced Panel: n = 50, T = 8, N = 400
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.10274 -1.55544 -0.28595 1.43194 14.06921
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
DDR -0.448772 0.045998 -9.7563 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 6788
Residual Sum of Squares: 5333.4
R-Squared: 0.21429
Adj. R-Squared: 0.10173
F-statistic: 95.1858 on 1 and 349 DF, p-value: < 2.22e-16summary(Drandom2)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ DDR, data = twoyeartestmerge, model = "random",
index = c("State", "year"))
Balanced Panel: n = 50, T = 8, N = 400
Effects:
var std.dev share
idiosyncratic 15.282 3.909 0.136
individual 97.086 9.853 0.864
theta: 0.8611
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.9272 -1.9749 -0.6160 1.5994 15.2403
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 66.414709 2.168685 30.6244 < 2.2e-16 ***
DDR -0.392888 0.045081 -8.7151 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 7561
Residual Sum of Squares: 6349.3
R-Squared: 0.16025
Adj. R-Squared: 0.15814
Chisq: 75.953 on 1 DF, p-value: < 2.22e-16phtest(Dfixed2, Drandom2)
Hausman Test
data: nmfr ~ DDR
chisq = 37.401, df = 1, p-value = 9.616e-10
alternative hypothesis: one model is inconsistentOPfixed2 <- plm(nmfr ~ OPDR, data=twoyeartestmerge, index=c("State", "year"), model="within")
OPrandom2 <- plm(nmfr ~ OPDR, data=twoyeartestmerge, index=c("State", "year"), model="random")
summary(OPfixed2)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ OPDR, data = twoyeartestmerge, model = "within",
index = c("State", "year"))
Balanced Panel: n = 50, T = 8, N = 400
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-48.96663 -1.68497 -0.29063 1.46926 14.36966
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
OPDR -0.473496 0.078315 -6.046 3.817e-09 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 6788
Residual Sum of Squares: 6144.5
R-Squared: 0.094811
Adj. R-Squared: -0.034873
F-statistic: 36.5547 on 1 and 349 DF, p-value: 3.817e-09summary(OPrandom2)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ OPDR, data = twoyeartestmerge, model = "random",
index = c("State", "year"))
Balanced Panel: n = 50, T = 8, N = 400
Effects:
var std.dev share
idiosyncratic 17.606 4.196 0.148
individual 101.725 10.086 0.852
theta: 0.8545
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-49.69446 -2.05139 -0.62305 1.55192 16.32835
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 56.325594 1.592988 35.3585 < 2.2e-16 ***
OPDR -0.457139 0.076321 -5.9897 2.102e-09 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 7636.3
Residual Sum of Squares: 7004.9
R-Squared: 0.082688
Adj. R-Squared: 0.080383
Chisq: 35.8763 on 1 DF, p-value: 2.1025e-09phtest(OPfixed2, OPrandom2)
Hausman Test
data: nmfr ~ OPDR
chisq = 0.86772, df = 1, p-value = 0.3516
alternative hypothesis: one model is inconsistent49. fixed effect models with states included with one year lags
MLmodel1 <- lm(nmfr ~ MLDR + State -1, data=oneyeartestmerge)
summary(MLmodel1)
Call:
lm(formula = nmfr ~ MLDR + State - 1, data = oneyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-7.0089 -1.7219 -0.4339 1.2604 13.6883
Coefficients:
Estimate Std. Error t value Pr(>|t|)
MLDR -0.09416 0.01619 -5.816 1.55e-08 ***
StateAlabama 101.59729 7.84140 12.957 < 2e-16 ***
StateAlaska 90.58841 5.73615 15.793 < 2e-16 ***
StateArizona 98.12316 5.57436 17.603 < 2e-16 ***
StateArkansas 114.58654 7.60529 15.067 < 2e-16 ***
StateCalifornia 81.40994 4.49992 18.091 < 2e-16 ***
StateColorado 63.27919 4.66801 13.556 < 2e-16 ***
StateConnecticut 60.68468 4.34900 13.954 < 2e-16 ***
StateDelaware 84.44474 5.90812 14.293 < 2e-16 ***
StateDistrict of Columbia 80.59649 7.12857 11.306 < 2e-16 ***
StateFlorida 89.45518 5.72731 15.619 < 2e-16 ***
StateGeorgia 96.35219 6.18325 15.583 < 2e-16 ***
StateHawaii 79.37662 4.71178 16.846 < 2e-16 ***
StateIdaho 76.78002 5.01633 15.306 < 2e-16 ***
StateIllinois 81.96931 5.13379 15.967 < 2e-16 ***
StateIndiana 95.66966 6.21531 15.393 < 2e-16 ***
StateIowa 79.12240 4.95214 15.977 < 2e-16 ***
StateKansas 90.19659 5.51866 16.344 < 2e-16 ***
StateKentucky 108.33021 7.52914 14.388 < 2e-16 ***
StateLouisiana 112.27426 7.54423 14.882 < 2e-16 ***
StateMaine 72.14438 5.08575 14.186 < 2e-16 ***
StateMaryland 74.66603 5.24981 14.223 < 2e-16 ***
StateMassachusetts 66.28542 4.44756 14.904 < 2e-16 ***
StateMichigan 82.99105 5.90108 14.064 < 2e-16 ***
StateMinnesota 67.53884 4.13814 16.321 < 2e-16 ***
StateMississippi 118.39074 8.31497 14.238 < 2e-16 ***
StateMissouri 92.54272 6.35614 14.560 < 2e-16 ***
StateMontana 83.10605 5.71756 14.535 < 2e-16 ***
StateNebraska 78.96201 4.82142 16.377 < 2e-16 ***
StateNevada 96.24353 6.00622 16.024 < 2e-16 ***
StateNew Hampshire 61.31299 4.54829 13.480 < 2e-16 ***
StateNew Jersey 65.19093 4.58885 14.206 < 2e-16 ***
StateNew Mexico 101.95220 6.56426 15.531 < 2e-16 ***
StateNew York 70.70534 4.55501 15.523 < 2e-16 ***
StateNorth Carolina 88.30275 5.96057 14.814 < 2e-16 ***
StateOhio 92.84848 6.19644 14.984 < 2e-16 ***
StateOklahoma 118.84206 7.67871 15.477 < 2e-16 ***
StateOregon 73.57615 5.11936 14.372 < 2e-16 ***
StatePennsylvania 77.97971 5.58251 13.969 < 2e-16 ***
StateRhode Island 67.75749 4.94043 13.715 < 2e-16 ***
StateSouth Carolina 100.09794 6.96033 14.381 < 2e-16 ***
StateSouth Dakota 89.99210 5.19653 17.318 < 2e-16 ***
StateTennessee 104.13442 7.33576 14.195 < 2e-16 ***
StateTexas 100.94570 5.56938 18.125 < 2e-16 ***
StateUtah 66.28520 4.78298 13.859 < 2e-16 ***
StateVermont 63.68725 4.54269 14.020 < 2e-16 ***
StateVirginia 78.98087 5.04225 15.664 < 2e-16 ***
StateWashington 71.45334 4.66711 15.310 < 2e-16 ***
StateWest Virginia 107.03932 8.01354 13.357 < 2e-16 ***
StateWisconsin 73.70766 4.77107 15.449 < 2e-16 ***
StateWyoming 86.09606 5.75611 14.957 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 2.994 on 299 degrees of freedom
Multiple R-squared: 0.9974, Adjusted R-squared: 0.9969
F-statistic: 2226 on 51 and 299 DF, p-value: < 2.2e-16Dmodel1 <- lm(nmfr ~ DDR + State -1, data=oneyeartestmerge)
summary(Dmodel1)
Call:
lm(formula = nmfr ~ DDR + State - 1, data = oneyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-7.5896 -1.4292 -0.4071 1.1550 12.8086
Coefficients:
Estimate Std. Error t value Pr(>|t|)
DDR -0.38663 0.03351 -11.54 <2e-16 ***
StateAlabama 68.89032 1.46467 47.03 <2e-16 ***
StateAlaska 78.97004 2.07993 37.97 <2e-16 ***
StateArizona 84.97976 1.89348 44.88 <2e-16 ***
StateArkansas 83.75382 1.49590 55.99 <2e-16 ***
StateCalifornia 68.09786 1.43918 47.32 <2e-16 ***
StateColorado 54.16669 1.79310 30.21 <2e-16 ***
StateConnecticut 47.20907 1.37347 34.37 <2e-16 ***
StateDelaware 64.38897 1.54591 41.65 <2e-16 ***
StateDistrict of Columbia 50.97908 1.39582 36.52 <2e-16 ***
StateFlorida 70.52121 1.54922 45.52 <2e-16 ***
StateGeorgia 71.87264 1.36918 52.49 <2e-16 ***
StateHawaii 63.72185 1.37347 46.40 <2e-16 ***
StateIdaho 64.51235 1.71656 37.58 <2e-16 ***
StateIllinois 62.93073 1.32300 47.57 <2e-16 ***
StateIndiana 73.91675 1.55400 47.57 <2e-16 ***
StateIowa 62.14011 1.38010 45.03 <2e-16 ***
StateKansas 71.19802 1.46431 48.62 <2e-16 ***
StateKentucky 82.64022 1.82029 45.40 <2e-16 ***
StateLouisiana 81.46931 1.47454 55.25 <2e-16 ***
StateMaine 57.12053 1.55547 36.72 <2e-16 ***
StateMaryland 55.37238 1.34828 41.07 <2e-16 ***
StateMassachusetts 52.82968 1.41069 37.45 <2e-16 ***
StateMichigan 63.05431 1.55105 40.65 <2e-16 ***
StateMinnesota 55.17557 1.36392 40.45 <2e-16 ***
StateMississippi 81.67029 1.38810 58.84 <2e-16 ***
StateMissouri 70.34930 1.58028 44.52 <2e-16 ***
StateMontana 69.22849 1.90204 36.40 <2e-16 ***
StateNebraska 61.54062 1.30887 47.02 <2e-16 ***
StateNevada 80.58865 1.89715 42.48 <2e-16 ***
StateNew Hampshire 51.06851 1.66197 30.73 <2e-16 ***
StateNew Jersey 48.48376 1.27131 38.14 <2e-16 ***
StateNew Mexico 89.61452 2.40452 37.27 <2e-16 ***
StateNew York 54.07978 1.26477 42.76 <2e-16 ***
StateNorth Carolina 66.38421 1.44543 45.93 <2e-16 ***
StateOhio 71.96390 1.60498 44.84 <2e-16 ***
StateOklahoma 93.03455 1.87599 49.59 <2e-16 ***
StateOregon 61.21293 1.75345 34.91 <2e-16 ***
StatePennsylvania 60.03786 1.55805 38.53 <2e-16 ***
StateRhode Island 54.88819 1.64284 33.41 <2e-16 ***
StateSouth Carolina 73.33326 1.51351 48.45 <2e-16 ***
StateSouth Dakota 74.74964 1.58513 47.16 <2e-16 ***
StateTennessee 77.22400 1.65277 46.72 <2e-16 ***
StateTexas 79.60773 1.33990 59.41 <2e-16 ***
StateUtah 57.09154 1.83717 31.08 <2e-16 ***
StateVermont 52.00721 1.56174 33.30 <2e-16 ***
StateVirginia 60.49947 1.32363 45.71 <2e-16 ***
StateWashington 60.06511 1.63219 36.80 <2e-16 ***
StateWest Virginia 81.56153 2.04763 39.83 <2e-16 ***
StateWisconsin 59.52044 1.48697 40.03 <2e-16 ***
StateWyoming 73.74622 2.03341 36.27 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 2.628 on 299 degrees of freedom
Multiple R-squared: 0.998, Adjusted R-squared: 0.9976
F-statistic: 2892 on 51 and 299 DF, p-value: < 2.2e-16OPmodel1 <- lm(nmfr ~ OPDR + State -1, data=oneyeartestmerge)
summary(OPmodel1)
Call:
lm(formula = nmfr ~ OPDR + State - 1, data = oneyeartestmerge)
Residuals:
Min 1Q Median 3Q Max
-7.0476 -1.6341 -0.3391 1.2784 13.2370
Coefficients:
Estimate Std. Error t value Pr(>|t|)
OPDR -0.3726 0.0588 -6.336 8.65e-10 ***
StateAlabama 58.1291 1.1514 50.487 < 2e-16 ***
StateAlaska 61.6877 1.2719 48.501 < 2e-16 ***
StateArizona 69.7784 1.2430 56.137 < 2e-16 ***
StateArkansas 73.2570 1.1840 61.875 < 2e-16 ***
StateCalifornia 57.9101 1.1578 50.018 < 2e-16 ***
StateColorado 39.8882 1.2139 32.860 < 2e-16 ***
StateConnecticut 40.0356 1.2695 31.536 < 2e-16 ***
StateDelaware 55.2216 1.3274 41.601 < 2e-16 ***
StateDistrict of Columbia 42.6318 1.2152 35.083 < 2e-16 ***
StateFlorida 59.7979 1.2168 49.143 < 2e-16 ***
StateGeorgia 63.2274 1.1751 53.806 < 2e-16 ***
StateHawaii 54.4779 1.1529 47.252 < 2e-16 ***
StateIdaho 50.1824 1.1576 43.351 < 2e-16 ***
StateIllinois 55.9376 1.2228 45.747 < 2e-16 ***
StateIndiana 62.3872 1.1766 53.023 < 2e-16 ***
StateIowa 53.1583 1.1680 45.511 < 2e-16 ***
StateKansas 60.7510 1.1634 52.217 < 2e-16 ***
StateKentucky 70.8337 1.4470 48.952 < 2e-16 ***
StateLouisiana 70.4736 1.1486 61.358 < 2e-16 ***
StateMaine 47.2082 1.2791 36.907 < 2e-16 ***
StateMaryland 49.3001 1.3229 37.266 < 2e-16 ***
StateMassachusetts 46.1448 1.3598 33.935 < 2e-16 ***
StateMichigan 52.6600 1.2398 42.476 < 2e-16 ***
StateMinnesota 46.4111 1.1657 39.814 < 2e-16 ***
StateMississippi 71.8693 1.1424 62.912 < 2e-16 ***
StateMissouri 60.0616 1.2787 46.971 < 2e-16 ***
StateMontana 52.9158 1.1837 44.704 < 2e-16 ***
StateNebraska 52.8427 1.1355 46.539 < 2e-16 ***
StateNevada 67.4719 1.4213 47.471 < 2e-16 ***
StateNew Hampshire 41.3814 1.4364 28.809 < 2e-16 ***
StateNew Jersey 41.5829 1.1764 35.349 < 2e-16 ***
StateNew Mexico 69.7638 1.4100 49.476 < 2e-16 ***
StateNew York 47.8593 1.2060 39.685 < 2e-16 ***
StateNorth Carolina 57.7257 1.2470 46.291 < 2e-16 ***
StateOhio 62.6209 1.3899 45.054 < 2e-16 ***
StateOklahoma 79.6834 1.3723 58.067 < 2e-16 ***
StateOregon 47.7848 1.2327 38.765 < 2e-16 ***
StatePennsylvania 48.9159 1.2011 40.726 < 2e-16 ***
StateRhode Island 45.5890 1.4475 31.495 < 2e-16 ***
StateSouth Carolina 62.7627 1.1943 52.551 < 2e-16 ***
StateSouth Dakota 62.0216 1.1468 54.084 < 2e-16 ***
StateTennessee 66.3069 1.3128 50.509 < 2e-16 ***
StateTexas 70.8794 1.1510 61.582 < 2e-16 ***
StateUtah 44.9413 1.4359 31.298 < 2e-16 ***
StateVermont 41.6390 1.2526 33.242 < 2e-16 ***
StateVirginia 53.1120 1.1999 44.264 < 2e-16 ***
StateWashington 48.6373 1.2511 38.875 < 2e-16 ***
StateWest Virginia 71.0116 1.9505 36.408 < 2e-16 ***
StateWisconsin 50.1570 1.2434 40.339 < 2e-16 ***
StateWyoming 56.2094 1.2132 46.330 < 2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 2.966 on 299 degrees of freedom
Multiple R-squared: 0.9974, Adjusted R-squared: 0.997
F-statistic: 2268 on 51 and 299 DF, p-value: < 2.2e-1650. fixed and random effect models with one year lags and Hausman test
MLfixed1 <- plm(nmfr ~ MLDR, data=oneyeartestmerge, index=c("State", "year"), model="within")
MLrandom1 <- plm(nmfr ~ MLDR, data=oneyeartestmerge, index=c("State", "year"), model="random")
summary(MLfixed1)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ MLDR, data = oneyeartestmerge, model = "within",
index = c("State", "year"))
Balanced Panel: n = 50, T = 7, N = 350
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-7.00892 -1.72194 -0.43388 1.26043 13.68834
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
MLDR -0.094161 0.016190 -5.8161 1.547e-08 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 2984
Residual Sum of Squares: 2680.8
R-Squared: 0.10164
Adj. R-Squared: -0.048593
F-statistic: 33.827 on 1 and 299 DF, p-value: 1.5469e-08summary(MLrandom1)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ MLDR, data = oneyeartestmerge, model = "random",
index = c("State", "year"))
Balanced Panel: n = 50, T = 7, N = 350
Effects:
var std.dev share
idiosyncratic 8.966 2.994 0.152
individual 49.937 7.067 0.848
theta: 0.8419
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-6.20990 -2.32082 -0.48372 1.66189 15.68283
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 47.370355 4.318366 10.9695 <2e-16 ***
MLDR 0.016124 0.012035 1.3397 0.1803
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 3882.6
Residual Sum of Squares: 3862.6
R-Squared: 0.005131
Adj. R-Squared: 0.0022722
Chisq: 1.79481 on 1 DF, p-value: 0.18034phtest(MLfixed1, MLrandom1)
Hausman Test
data: nmfr ~ MLDR
chisq = 103.73, df = 1, p-value < 2.2e-16
alternative hypothesis: one model is inconsistentDfixed1 <- plm(nmfr ~ DDR, data=oneyeartestmerge, index=c("State", "year"), model="within")
Drandom1 <- plm(nmfr ~ DDR, data=oneyeartestmerge, index=c("State", "year"), model="random")
summary(Dfixed1)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ DDR, data = oneyeartestmerge, model = "within",
index = c("State", "year"))
Balanced Panel: n = 50, T = 7, N = 350
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-7.58959 -1.42918 -0.40709 1.15496 12.80861
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
DDR -0.386630 0.033505 -11.539 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 2984
Residual Sum of Squares: 2064.6
R-Squared: 0.30812
Adj. R-Squared: 0.19242
F-statistic: 133.156 on 1 and 299 DF, p-value: < 2.22e-16summary(Drandom1)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ DDR, data = oneyeartestmerge, model = "random",
index = c("State", "year"))
Balanced Panel: n = 50, T = 7, N = 350
Effects:
var std.dev share
idiosyncratic 6.905 2.628 0.064
individual 100.860 10.043 0.936
theta: 0.9016
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-5.60097 -1.78912 -0.39454 1.25752 14.09409
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 66.076374 1.900229 34.773 < 2.2e-16 ***
DDR -0.359431 0.033442 -10.748 < 2.2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 3332.1
Residual Sum of Squares: 2501.6
R-Squared: 0.24922
Adj. R-Squared: 0.24706
Chisq: 115.519 on 1 DF, p-value: < 2.22e-16phtest(Dfixed1, Drandom1)
Hausman Test
data: nmfr ~ DDR
chisq = 173.57, df = 1, p-value < 2.2e-16
alternative hypothesis: one model is inconsistentOPfixed1 <- plm(nmfr ~ OPDR, data=oneyeartestmerge, index=c("State", "year"), model="within")
OPrandom1 <- plm(nmfr ~ OPDR, data=oneyeartestmerge, index=c("State", "year"), model="random")
summary(OPfixed1)Oneway (individual) effect Within Model
Call:
plm(formula = nmfr ~ OPDR, data = oneyeartestmerge, model = "within",
index = c("State", "year"))
Balanced Panel: n = 50, T = 7, N = 350
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-7.04761 -1.63409 -0.33914 1.27841 13.23702
Coefficients:
Estimate Std. Error t-value Pr(>|t|)
OPDR -0.372572 0.058801 -6.3361 8.648e-10 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 2984
Residual Sum of Squares: 2630.8
R-Squared: 0.11838
Adj. R-Squared: -0.029054
F-statistic: 40.1465 on 1 and 299 DF, p-value: 8.6476e-10summary(OPrandom1)Oneway (individual) effect Random Effect Model
(Swamy-Arora's transformation)
Call:
plm(formula = nmfr ~ OPDR, data = oneyeartestmerge, model = "random",
index = c("State", "year"))
Balanced Panel: n = 50, T = 7, N = 350
Effects:
var std.dev share
idiosyncratic 8.799 2.966 0.077
individual 104.991 10.247 0.923
theta: 0.8912
Residuals:
Min. 1st Qu. Median 3rd Qu. Max.
-5.33229 -1.97631 -0.50823 1.38051 14.81980
Coefficients:
Estimate Std. Error z-value Pr(>|z|)
(Intercept) 56.293865 1.548441 36.3552 < 2.2e-16 ***
OPDR -0.366979 0.057854 -6.3432 2.25e-10 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Total Sum of Squares: 3409.1
Residual Sum of Squares: 3055.8
R-Squared: 0.10364
Adj. R-Squared: 0.10106
Chisq: 40.2363 on 1 DF, p-value: 2.2502e-10phtest(OPfixed1, OPrandom1)
Hausman Test
data: nmfr ~ OPDR
chisq = 0.28305, df = 1, p-value = 0.5947
alternative hypothesis: one model is inconsistentWorks Cited:
Case, A., & Deaton, A. (2017). Mortality and morbidity in the 21st century. Brookings papers on economic activity, 397.
Caudillo, M. L., & Villarreal, A. (2021). The Opioid Epidemic and Nonmarital Childbearing in the United States, 2000–2016. Demography, 58(1), 345-378.
Bachrach, C. A., & Morgan, S. P. (2013). A cognitive–social model of fertility intentions. Population and development review, 39(3), 459-485.
Kearney, M. S., & Levine, P. B. (2011). Income Inequality and Early Non-Marital Childbearing: An Economic Exploration of the” Culture of Despair” (No. w17157). National Bureau of economic research.
Kearney, M. S., & Levine, P. B. (2015). Investigating recent trends in the US teen birth rate. Journal of Health Economics, 41, 15-29.
Lutz, W., Skirbekk, V., & Testa, M. R. (2006). The low-fertility trap hypothesis: Forces that may lead to further postponement and fewer births in Europe. Vienna yearbook of population research, 167-192.
Melchior, M., Berkman, L. F., Kawachi, I., Krieger, N., Zins, M., Bonenfant, S., & Goldberg, M. (2006). Lifelong socioeconomic trajectory and premature mortality (35–65 years) in France: findings from the GAZEL Cohort Study. Journal of Epidemiology & Community Health, 60(11), 937-944.
Merton, Robert K. 1968. “The Matthew Effect in Science.” Science 159(3810):56–63.
Montez, J. K., & Hayward, M. D. (2014). Cumulative childhood adversity, educational attainment, and active life expectancy among US adults. Demography, 51(2), 413-435.
Montez JK,Mehri N,Monnat SM, Beckfield J,Chapman D,Grumbach JM,etal. (2022) U.S.statepolicy contexts andmortality of working-ageadults. PLoS ONE17(10): e0275466. https://doi.org/10.1371/journal.pone.0275466
Schneider, D., & Gemmill, A. (2016). The Surprising Decline in the Non-Marital Fertility Rate in the United States. Population and Development Review, 42(4), 627–649. http://www.jstor.org/stable/44132227
Woolf SH, Schoomaker H. Life Expectancy and Mortality Rates in the United States, 1959-2017. JAMA. 2019;322(20):1996–2016. doi:10.1001/jama.2019.16932
Woolf S H, Chapman D A, Buchanich J M, Bobby K J, Zimmerman E B, Blackburn S M et al. Changes in midlife death rates across racial and ethnic groups in the United States: systematic analysis of vital statistics BMJ 2018; 362 :k3096 doi:10.1136/bmj.k3096