Şimdiye kadar R studio’da öğrendiğimiz ve uygulandığımız dersler ayrıltılı bir şekilde anlatmaya çalışırız.Bunun için sırayla ve kademe kademe işletiriz.
R MARKDOWN
Herhangi R’da bir veri seti seçtigimiz zaman R MARKDOWN ve RMARKDOWN paketi ’page_table’kullanarak verilerimiz tablolaştırabiliriz.
WOOLDRİDGE
O veriler uygulanmak için hazırlanmış bir pakettir.
Library(), Require()
Yüklenen paketler aktif haleye getirmek ve R programına hatırlamak için Library veya Require komutları kullanılır.
Şimdi pratiğe geçelim et tüm bunlar uygulayalım. Öncelikle bir veri seti seçelim. veri setimiz “alcohol”
library(wooldridge)
library(rmarkdown)
data("alcohol")
paged_table(alcohol)Bir data verileri çekmek için iki yöntem vardır. İlk yöntem:
lm(abuse~status+unemrate+age+educ+married+famsize+white+exhealth,data = alcohol)##
## Call:
## lm(formula = abuse ~ status + unemrate + age + educ + married +
## famsize + white + exhealth, data = alcohol)
##
## Coefficients:
## (Intercept) status unemrate age educ married
## 2.152e-01 -7.905e-03 -2.274e-03 1.518e-05 -3.496e-03 -8.039e-03
## famsize white exhealth
## -1.304e-02 1.975e-02 -2.297e-02Ve ikinci yöntemimiz:
lm(formula = alcohol$abuse~alcohol$status+alcohol$unemrate+alcohol$age+alcohol$educ+alcohol$famsize+alcohol$white+alcohol$exhealth)##
## Call:
## lm(formula = alcohol$abuse ~ alcohol$status + alcohol$unemrate +
## alcohol$age + alcohol$educ + alcohol$famsize + alcohol$white +
## alcohol$exhealth)
##
## Coefficients:
## (Intercept) alcohol$status alcohol$unemrate alcohol$age
## 2.163e-01 -8.332e-03 -2.362e-03 -7.819e-05
## alcohol$educ alcohol$famsize alcohol$white alcohol$exhealth
## -3.466e-03 -1.387e-02 1.927e-02 -2.315e-02Birinci yada ikinci hangisi kullanırsak olsun sonuçlar aynıdır.
SUMMARY
Veriler kolay anlaşılmasını ve özetlemesine summary kumutu kullanıyoruz. Hemen pratiğe geçelim.
summary(lm(formula = abuse~status+unemrate+age+educ+married+famsize+white+exhealth, data=alcohol))##
## Call:
## lm(formula = abuse ~ status + unemrate + age + educ + married +
## famsize + white + exhealth, data = alcohol)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.1935 -0.1154 -0.0957 -0.0713 0.9961
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.152e-01 2.863e-02 7.516 6.14e-14 ***
## status -7.905e-03 5.890e-03 -1.342 0.179610
## unemrate -2.274e-03 2.004e-03 -1.135 0.256500
## age 1.518e-05 3.330e-04 0.046 0.963650
## educ -3.496e-03 1.079e-03 -3.239 0.001203 **
## married -8.039e-03 8.975e-03 -0.896 0.370477
## famsize -1.304e-02 2.187e-03 -5.964 2.55e-09 ***
## white 1.975e-02 8.634e-03 2.288 0.022173 *
## exhealth -2.297e-02 6.280e-03 -3.658 0.000256 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2977 on 9813 degrees of freedom
## Multiple R-squared: 0.008511, Adjusted R-squared: 0.007702
## F-statistic: 10.53 on 8 and 9813 DF, p-value: 8.005e-15STARGAZER
Stargazer, sonuçlarımızı daha açıklayıcı tablolarda gösterbilmemizi veya bir ve birden fazla değişkenle karşılaştırbilmemizi sağlar. Farklı regresyon tablolarını bir ararda göstermek için Stargazer paketini kullanırız.
model1<-lm(abuse~I(status*2)+unemrate+age+educ+married+famsize+white+exhealth,data = alcohol)
model2<-lm(abuse~status+unemrate+I(age/2)+educ+married+famsize+white+exhealth,data = alcohol)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=stargazerstargazer(list(model1,model2),type = "text")##
## ============================================================
## Dependent variable:
## ----------------------------
## abuse
## (1) (2)
## ------------------------------------------------------------
## I(status * 2) -0.004
## (0.003)
##
## status -0.008
## (0.006)
##
## unemrate -0.002 -0.002
## (0.002) (0.002)
##
## age 0.00002
## (0.0003)
##
## I(age/2) 0.00003
## (0.001)
##
## educ -0.003*** -0.003***
## (0.001) (0.001)
##
## married -0.008 -0.008
## (0.009) (0.009)
##
## famsize -0.013*** -0.013***
## (0.002) (0.002)
##
## white 0.020** 0.020**
## (0.009) (0.009)
##
## exhealth -0.023*** -0.023***
## (0.006) (0.006)
##
## Constant 0.215*** 0.215***
## (0.029) (0.029)
##
## ------------------------------------------------------------
## Observations 9,822 9,822
## R2 0.009 0.009
## Adjusted R2 0.008 0.008
## Residual Std. Error (df = 9813) 0.298 0.298
## F Statistic (df = 8; 9813) 10.529*** 10.529***
## ============================================================
## Note: *p<0.1; **p<0.05; ***p<0.01status değeri: -0.004
age değeri: 0.00003
STANDARDİZASYONlm(scale(abuse)~scale(status)+scale(unemrate)+scale(age)+scale(educ)+scale(married)+scale(famsize)+scale(white)+scale(exhealth),data = alcohol)##
## Call:
## lm(formula = scale(abuse) ~ scale(status) + scale(unemrate) +
## scale(age) + scale(educ) + scale(married) + scale(famsize) +
## scale(white) + scale(exhealth), data = alcohol)
##
## Coefficients:
## (Intercept) scale(status) scale(unemrate) scale(age)
## 1.252e-14 -1.403e-02 -1.145e-02 4.894e-04
## scale(educ) scale(married) scale(famsize) scale(white)
## -3.391e-02 -1.041e-02 -6.651e-02 2.340e-02
## scale(exhealth)
## -3.788e-02DPLYR
Dplyr paketi veriişlemenin grameri olarak bilinir.6 ayrı fonksiyona sahip olduğu için bir sürü değişim,düzenleme vs...yardımcı oluyor. Bunlar:Select(),Filter(),Mutate(),Group_by()ve son olarak Summarize()
Şimdi verilemiz hazırlayalım.library(wooldridge)
library(rmarkdown)
data("alcohol")
paged_table(alcohol)Dplyr'imiz indirdik,veri setimizde değişiklikler yapalım.require(dplyr)## Loading required package: 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, unionalcohol%>%
group_by(educ)%>%
summarise("ortalama büyüklük(unemrate)"=mean(unemrate))## # A tibble: 20 × 2
## educ `ortalama büyüklük(unemrate)`
## <int> <dbl>
## 1 0 6.02
## 2 1 5.79
## 3 2 6.07
## 4 3 5.78
## 5 4 5.86
## 6 5 5.71
## 7 6 5.55
## 8 7 5.73
## 9 8 5.53
## 10 9 5.62
## 11 10 5.71
## 12 11 5.82
## 13 12 5.56
## 14 13 5.64
## 15 14 5.54
## 16 15 5.76
## 17 16 5.47
## 18 17 5.51
## 19 18 5.51
## 20 19 5.62Bütün değişkenlerin yaş’a göre ortalamısını görmek için de across(everything())komutunu kullanabiliriz.
require(dplyr)
alcohol%>%
group_by(age) %>%
summarize(across(everything(),mean))## # A tibble: 35 × 33
## age abuse status unemrate educ married famsize white exhealth vghealth
## <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 25 0.126 2.82 5.54 13.3 0.489 2.16 0.861 0.434 0.344
## 2 26 0.1 2.86 5.54 13.2 0.578 2.22 0.867 0.453 0.325
## 3 27 0.0815 2.86 5.53 13.2 0.614 2.43 0.834 0.497 0.307
## 4 28 0.0948 2.85 5.61 13.3 0.641 2.43 0.855 0.456 0.302
## 5 29 0.0894 2.85 5.54 13.5 0.709 2.65 0.849 0.453 0.341
## 6 30 0.0921 2.85 5.55 13.4 0.710 2.78 0.832 0.461 0.344
## 7 31 0.0823 2.88 5.61 13.5 0.769 2.80 0.835 0.468 0.293
## 8 32 0.0865 2.89 5.59 13.6 0.749 2.74 0.896 0.539 0.268
## 9 33 0.0877 2.88 5.60 13.4 0.777 2.98 0.832 0.436 0.338
## 10 34 0.111 2.87 5.48 13.7 0.785 2.93 0.848 0.459 0.299
## # … with 25 more rows, and 23 more variables: goodhealth <dbl>,
## # fairhealth <dbl>, northeast <dbl>, midwest <dbl>, south <dbl>,
## # centcity <dbl>, outercity <dbl>, qrt1 <dbl>, qrt2 <dbl>, qrt3 <dbl>,
## # beertax <dbl>, cigtax <dbl>, ethanol <dbl>, mothalc <dbl>, fathalc <dbl>,
## # livealc <dbl>, inwf <dbl>, employ <dbl>, agesq <dbl>, beertaxsq <dbl>,
## # cigtaxsq <dbl>, ethanolsq <dbl>, educsq <dbl>