library(wooldridge)## Warning: package 'wooldridge' was built under R version 4.1.3library(rmarkdown)data("murder")paged_table(murder)##Panel veri setisi oluşturma(index)
library(plm)## Warning: package 'plm' was built under R version 4.1.3murder.yahya1<- pdata.frame(murder, index = c("state", "year"))pdim(murder.yahya1)## Balanced Panel: n = 51, T = 3, N = 153##1A: Cinayet mahkumlarının geçmişteki infazları caydırıcı bir etkiye sahipse b1 ne olur? b2’nin sahip olması gereken işaret hakkında ne düşünülüyorsünuz?
model1<- plm(mrdrte~exec+unem, data = murder.yahya1, model = "within")summary(model1)## Oneway (individual) effect Within Model
##
## Call:
## plm(formula = mrdrte ~ exec + unem, data = murder.yahya1, model = "within")
##
## Balanced Panel: n = 51, T = 3, N = 153
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -27.886739 -0.473005 0.059857 0.594079 14.202250
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## exec -0.114074 0.180084 -0.6335 0.5279
## unem 0.095914 0.280072 0.3425 0.7327
##
## Total Sum of Squares: 1311.5
## Residual Sum of Squares: 1305.3
## R-Squared: 0.0046814
## Adj. R-Squared: -0.51288
## F-statistic: 0.235172 on 2 and 100 DF, p-value: 0.79087infaz oranı artarsa cinayet oranı -0.114 azalmaktadır ancak unem oranı artarsa cinayet oranı 0.095 artıyor demektir
##1B. sadece 1990 ve 1993 yillarını kullanarak soru A’deki eşitliği havuzlanmış SEKK ile tahmin ediniz.
require(dplyr)## Loading required package: dplyr##
## Attaching package: 'dplyr'## The following objects are masked from 'package:plm':
##
## between, lag, lead## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionmodel_90_93<- filter(murder.yahya1, year %in% c(90,93))head(model_90_93)## id state year mrdrte exec unem d90 d93 cmrdrte cexec cunem
## AK-90 2 AK 90 7.5 0 6.9 1 0 -2.6000004 0 -3.9000001
## AK-93 2 AK 93 9.0 0 7.6 0 1 1.5000000 0 0.6999998
## AL-90 1 AL 90 11.6 5 6.8 1 0 2.3000002 3 -1.0000000
## AL-93 1 AL 93 11.6 2 7.5 0 1 0.0000000 -3 0.6999998
## AR-90 4 AR 90 10.3 2 6.9 1 0 2.7000003 2 -1.2000003
## AR-93 4 AR 93 10.2 2 6.2 0 1 -0.1000004 0 -0.7000003
## cexec_1 cunem_1
## AK-90 NA NA
## AK-93 0 -3.9
## AL-90 NA NA
## AL-93 3 -1.0
## AR-90 NA NA
## AR-93 2 -1.2model2<- plm(mrdrte~exec+unem, data = model_90_93, model = "pooling")summary(model2)## Pooling Model
##
## Call:
## plm(formula = mrdrte ~ exec + unem, data = model_90_93, model = "pooling")
##
## Balanced Panel: n = 34, T = 3, N = 102
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -12.91596 -3.39376 -1.17720 0.81388 67.59155
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## (Intercept) -4.88906 4.40781 -1.1092 0.270039
## exec 0.11491 0.26281 0.4372 0.662890
## unem 2.28750 0.74027 3.0901 0.002598 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 11401
## Residual Sum of Squares: 10340
## R-Squared: 0.093093
## Adj. R-Squared: 0.074771
## F-statistic: 5.0811 on 2 and 99 DF, p-value: 0.0079316(Intercept) -4.88906 burda negatif bir değer olduğundan dılayı (infaz ve işsizlik) değerlei anlamsız gösterşlmektedir
##1C. 1990 ve 1993’ü kullanarak eşitliği sabit etkilere tahmin ediniz.sadece iki yila aıt verileri kullanmanız nadeniyle ilk farkları kullanabilirsiniz.
library(plm)model3<- plm(mrdrte~exec+unem - 1, data = murder.yahya1, model = "fd")summary(model3)## Oneway (individual) effect First-Difference Model
##
## Call:
## plm(formula = mrdrte ~ exec + unem - 1, data = murder.yahya1,
## model = "fd")
##
## Balanced Panel: n = 51, T = 3, N = 153
## Observations used in estimation: 102
##
## Residuals:
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -2.788 -0.384 0.425 0.863 1.289 41.614
##
## Coefficients:
## Estimate Std. Error t-value Pr(>|t|)
## exec -0.11240 0.14926 -0.7530 0.4532
## unem -0.04823 0.26614 -0.1812 0.8566
##
## Total Sum of Squares: 1859
## Residual Sum of Squares: 1918.7
## R-Squared: 0.0089367
## Adj. R-Squared: -0.00097398
## F-statistic: 0.329508 on 2 and 100 DF, p-value: 0.7200614.10
library(plm)
data("airfare")library(rmarkdown)
paged_table(airfare)idata <- pdata.frame(airfare, index = c("id", "year"))pdim(idata)## Balanced Panel: n = 1149, T = 4, N = 4596pvar(idata)## no time variation: id dist ldist ldistsq
## no individual variation: year y98 y99 y00pooling <- plm(lpassen ~ fare + dist + passen + bmktshr + I(ldist^2) + year, data = idata, model = "pooling")within <- plm(lpassen ~ fare + dist + passen + bmktshr + I(ldist^2) + year, data = idata, model = "within")random <- plm(lpassen ~ fare + dist + passen + bmktshr + I(ldist^2) + year, data = idata, model = "random")summary(random)## Oneway (individual) effect Random Effect Model
## (Swamy-Arora's transformation)
##
## Call:
## plm(formula = lpassen ~ fare + dist + passen + bmktshr + I(ldist^2) +
## year, data = idata, model = "random")
##
## Balanced Panel: n = 1149, T = 4, N = 4596
##
## Effects:
## var std.dev share
## idiosyncratic 0.02046 0.14304 0.081
## individual 0.23084 0.48046 0.919
## theta: 0.8528
##
## Residuals:
## Min. 1st Qu. Median 3rd Qu. Max.
## -2.561815 -0.055653 0.015611 0.070798 1.641581
##
## Coefficients:
## Estimate Std. Error z-value Pr(>|z|)
## (Intercept) 5.8578e+00 1.9585e-01 29.9097 < 2.2e-16 ***
## fare -3.2722e-03 1.2039e-04 -27.1795 < 2.2e-16 ***
## dist 3.0581e-04 8.4715e-05 3.6099 0.0003063 ***
## passen 7.6429e-04 1.4579e-05 52.4224 < 2.2e-16 ***
## bmktshr 1.9990e-01 3.7525e-02 5.3271 9.977e-08 ***
## I(ldist^2) -4.3679e-03 5.9083e-03 -0.7393 0.4597420
## year1998 9.5304e-03 6.1333e-03 1.5539 0.1202158
## year1999 3.3770e-02 6.2124e-03 5.4358 5.454e-08 ***
## year2000 7.9356e-02 6.5168e-03 12.1771 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Total Sum of Squares: 203.17
## Residual Sum of Squares: 98.487
## R-Squared: 0.51524
## Adj. R-Squared: 0.51439
## Chisq: 4875.39 on 8 DF, p-value: < 2.22e-16library(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=stargazerstargazer(pooling, within, random, type = "text", column.labels = c("OLS","RE","FE"))##
## ===============================================================================
## Dependent variable:
## ------------------------------------------------------------------
## lpassen
## OLS RE FE
## (1) (2) (3)
## -------------------------------------------------------------------------------
## fare -0.001*** -0.004*** -0.003***
## (0.0001) (0.0001) (0.0001)
##
## dist 0.0001** 0.0003***
## (0.00004) (0.0001)
##
## passen 0.001*** 0.001*** 0.001***
## (0.00001) (0.00002) (0.00001)
##
## bmktshr 0.223*** 0.161*** 0.200***
## (0.047) (0.040) (0.038)
##
## I(ldist2) 0.001 -0.004
## (0.003) (0.006)
##
## year1998 0.004 0.015** 0.010
## (0.021) (0.006) (0.006)
##
## year1999 0.018 0.051*** 0.034***
## (0.021) (0.006) (0.006)
##
## year2000 0.041* 0.110*** 0.079***
## (0.021) (0.007) (0.007)
##
## Constant 5.370*** 5.858***
## (0.110) (0.196)
##
## -------------------------------------------------------------------------------
## Observations 4,596 4,596 4,596
## R2 0.677 0.450 0.515
## Adjusted R2 0.676 0.266 0.514
## F Statistic 1,199.361*** (df = 8; 4587) 469.900*** (df = 6; 3441) 4,875.385***
## ===============================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01phtest(pooling, random)##
## Hausman Test
##
## data: lpassen ~ fare + dist + passen + bmktshr + I(ldist^2) + year
## chisq = 1086.5, df = 8, p-value < 2.2e-16
## alternative hypothesis: one model is inconsistent