library(readr)
Evidencia_eco <- read_csv("C:/Users/Fercs/OneDrive/Tec de Monterrey/Uni/S4/Econometria/Evidencia_eco.csv")
## Rows: 9 Columns: 43
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (2): Location, Indicator
## num (41): 1980, 1981, 1982, 1983, 1984, 1985, 1986, 1987, 1988, 1989, 1990, ...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
DATOS<-Evidencia_eco
Limpieza de base
library(dplyr)
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
#Arreglo de columnas y variables
DATOS <- t(DATOS)
DATOS1 <- DATOS[-1, ]
nom_var<- DATOS1[1, ]
colnames(DATOS1) <- nom_var
DATOSF <- DATOS1[-1, ]
DATOSF_1 <- as.data.frame(DATOSF)
# Limpieza
convertir_numerico <- function(x) {
x <- as.numeric(gsub(",", "", x))
return(x)
}
# Aplicar la función a las columnas problemáticas
DATOSF_1$empleo.total <- convertir_numerico(DATOSF_1$empleo.total)
DATOSF_1$inversion.US <- convertir_numerico(DATOSF_1$inversion.US)
DATOSF_1$PIB.real <- convertir_numerico(DATOSF_1$PIB.real)
DATOSF_1$inversion.real <- convertir_numerico(DATOSF_1$inversion.real)
# Convertir las demás columnas a numérico
DATOSF_1$poblacion.total <- as.numeric(DATOSF_1$poblacion.total)
DATOSF_1$poblacion.crece <- as.numeric(DATOSF_1$poblacion.crece)
DATOSF_1$tasa.desempleo <- as.numeric(DATOSF_1$tasa.desempleo)
DATOSF_1$inflacion <- as.numeric(DATOSF_1$inflacion)
# Asignar el resultado a Data_final
data_final <- DATOSF_1
Modelo general
#PASO 1 : serie de tiempo
library(lubridate)
##
## Attaching package: 'lubridate'
## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, union
library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ purrr 1.0.2 ✔ tidyr 1.3.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(stargazer)
##
## Please cite as:
##
## Hlavac, Marek (2022). stargazer: Well-Formatted Regression and Summary Statistics Tables.
## R package version 5.2.3. https://CRAN.R-project.org/package=stargazer
DATOSF.ts <- ts(data_final, start = 1980, frequency = 1)
#Variables a ts
fecha.ts=ts(DATOSF.ts[,1],start=1980,frequency=1)
poblacion.total.ts=ts(DATOSF.ts[,2],start=1980,frequency=1)
poblacion.crece.ts=ts(DATOSF.ts[,3],start=1980,frequency=1)
empleo.total.ts=ts(DATOSF.ts[,4],start=1980,frequency=1)
inversion.US.ts=ts(DATOSF.ts[,5],start=1980,frequency=1)
PIB.real.ts=ts(DATOSF.ts[,6],start=1980,frequency=1)
inversion.real.ts=ts(DATOSF.ts[,7],start=1980,frequency=1)
tasa.desempleo.ts=ts(DATOSF.ts[,8],start=1980,frequency=1)
inflacion.ts=ts(DATOSF.ts[,9],start=1980,frequency=1)
#PASO 2: modelo regresión lineal
Model_general <- lm(tasa.desempleo.ts ~ poblacion.total.ts + poblacion.crece.ts + empleo.total.ts + inversion.US.ts + PIB.real.ts + inversion.real.ts + inflacion.ts)
stargazer(Model_general,type="text")
##
## ===============================================
## Dependent variable:
## ---------------------------
## tasa.desempleo.ts
## -----------------------------------------------
## poblacion.total.ts 0.403***
## (0.077)
##
## poblacion.crece.ts 2.738**
## (1.040)
##
## empleo.total.ts -0.0004***
## (0.0001)
##
## inversion.US.ts -0.00002
## (0.00002)
##
## PIB.real.ts -0.0005
## (0.0003)
##
## inversion.real.ts 0.001
## (0.0005)
##
## inflacion.ts 0.013**
## (0.005)
##
## Constant -17.335***
## (4.503)
##
## -----------------------------------------------
## Observations 41
## R2 0.549
## Adjusted R2 0.454
## Residual Std. Error 0.671 (df = 33)
## F Statistic 5.746*** (df = 7; 33)
## ===============================================
## Note: *p<0.1; **p<0.05; ***p<0.01
Modelo óptimo
# df para matriz de correlación
datos_ts <- data.frame(poblacion.total.ts,
poblacion.crece.ts,
empleo.total.ts,
inversion.US.ts,
PIB.real.ts,
inversion.real.ts,
tasa.desempleo.ts,
inflacion.ts)
Multicolinealidad
# Análisis de multicolinealidad
library(car)
## Loading required package: carData
##
## Attaching package: 'car'
## The following object is masked from 'package:purrr':
##
## some
## The following object is masked from 'package:dplyr':
##
## recode
library(corrplot)
## corrplot 0.92 loaded
library(dplyr)
# VIF
V1 <- vif(Model_general)
barplot(V1, main = "VIF", horiz = TRUE, col = "pink",cex.names=0.8,las=1)
abline(v = 10, lwd = 3, lty = 2)
# Matriz de correlación
corrM<-round(cor(datos_ts),2)
corrplot(corrM,method="number",type="upper",tl.col="black",tl.cex=0.5,number.digits=1,number.cex=0.5)
# Modelo tras haber corregido multicolinealidad
Model_op2 <- lm(tasa.desempleo.ts ~ poblacion.crece.ts + inversion.US.ts + inversion.real.ts + inflacion.ts)
vif(Model_op2)
## poblacion.crece.ts inversion.US.ts inversion.real.ts inflacion.ts
## 4.041619 3.045987 5.437193 2.463518
Autocorrelación
# Análisis de autocorrelación
#Durbin - Watson
library(lmtest)
## Loading required package: zoo
##
## Attaching package: 'zoo'
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
dwtest(Model_op2)
##
## Durbin-Watson test
##
## data: Model_op2
## DW = 0.86827, p-value = 3.434e-06
## alternative hypothesis: true autocorrelation is greater than 0
# Corrección por Prais-Winsten
library(prais)
## Loading required package: sandwich
## Loading required package: pcse
##
## Attaching package: 'pcse'
## The following object is masked from 'package:sandwich':
##
## vcovPC
Model_op22 <- prais_winsten(Model_op2,index=fecha.ts,data=DATOSF.ts)
## Iteration 0: rho = 0
## Iteration 1: rho = 0.5611
## Iteration 2: rho = 0.7944
## Iteration 3: rho = 0.9218
## Iteration 4: rho = 0.9805
## Iteration 5: rho = 0.9869
## Iteration 6: rho = 0.9865
## Iteration 7: rho = 0.9866
## Iteration 8: rho = 0.9866
## Iteration 9: rho = 0.9866
summary(Model_op22)
##
## Call:
## prais_winsten(formula = Model_op2, data = DATOSF.ts, index = fecha.ts)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.819 -2.735 -1.468 2.219 2.836
##
## AR(1) coefficient rho after 9 iterations: 0.9866
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.478e+00 2.865e+00 2.261 0.0299 *
## poblacion.crece.ts 1.579e+00 8.185e-01 1.929 0.0616 .
## inversion.US.ts -6.696e-06 1.005e-05 -0.666 0.5095
## inversion.real.ts -1.205e-03 2.131e-04 -5.654 2.02e-06 ***
## inflacion.ts 8.339e-04 3.638e-03 0.229 0.8200
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4691 on 36 degrees of freedom
## Multiple R-squared: 0.5329, Adjusted R-squared: 0.481
## F-statistic: 10.27 on 4 and 36 DF, p-value: 1.188e-05
##
## Durbin-Watson statistic (original): 0.8683
## Durbin-Watson statistic (transformed): 1.07