2
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 5 years have passed since last update.

おにぎり万太郎Advent Calendar 2019

Day 9

Linear Probability , Logit and Probit Model

Last updated at Posted at 2019-12-03

女性が働いている確率を説明するに当たって、outcome は働いているか、働いていないかというbinary responseであるという特徴から、タイトルの3つのモデルを考えます。
使うデータセットの説明は以下の通りです。

Wooldridge の Introductory Econometrics の公開されているデータセットの中のmrozというデータの内容で、

library(foreign)
mroz<-read.dta("http://fmwww.bc.edu/ec-p/data/wooldridge/mroz.dta")

というようにしてRで読み込めます。
使う変数は
inlf: in the labor force (働いているかどうか) nwifeinc: non wife income (自分で稼ぐ分以外の収入。夫の収入など) educ: education (教育年数) exper: experience(就業経験) age: age (年齢) kidslt6: kids less than 6 (6歳未満の子供の数) kidsge6: kids greater than or equal to 6 (6歳より上の歳の子供の数)

LPM , Logit , Probit regression (後者2つはMaximum Liklihood estimation)

LPM (Linear Probability Model)

$$inlf_i=\beta_0+\beta_1nwifeinc_i+\beta_2educ_i+\beta_3exper_i+\beta_4exper^2_i+\beta_5age_i+\beta_6kidslt6_i+\beta_7kidsge6_i+u_i$$
inlfがbinary(0,1変数)の時、OLSの回帰直線に当たるE(inlf)はinlfが1になる確率に一致する。

linprob <- lm(inlf~nwifeinc+educ+exper+I(exper^2)+age+kidslt6+kidsge6,data=mroz)

t-test using heteroscedasticity-robust SE(homoskedasticには構造上なり得ないので、今回はheteroskedsasticity-robust se を使ってt-test)


library(lmtest);library(car) 
coeftest(linprob,vcov=hccm)
t test of coefficients:
               Estimate  Std. Error t value  Pr(>|t|)    
(Intercept)  0.58551922  0.15358032  3.8125  0.000149 ***
nwifeinc    -0.00340517  0.00155826 -2.1852  0.029182 *  
educ         0.03799530  0.00733982  5.1766 2.909e-07 ***
exper        0.03949239  0.00598359  6.6001 7.800e-11 ***
I(exper^2)  -0.00059631  0.00019895 -2.9973  0.002814 ** 
age         -0.01609081  0.00241459 -6.6640 5.183e-11 ***
kidslt6     -0.26181047  0.03215160 -8.1430 1.621e-15 ***
kidsge6      0.01301223  0.01366031  0.9526  0.341123    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1`

どの変数のcoefficientも5%では有意。

例として2つの極端な値で働く確率を予測してみます。

prediction for 2 extreme women

xpred <- list(nwifeinc=c(100,0),educ=c(5,17),exper=c(0,30),
              age=c(20,52),kidslt6=c(2,0),kidsge6=c(0,0))
predict(linprob,xpred,type = "response")
         1          2 
-0.4104582  1.0428084 

####LPM の問題としては、回帰線は確率を表すのに1を超えたり0を下回ったりするところ。observaitionが端の方では合わない。この欠点を次の2つで克服しにいきます。

Logit model

Logistic distributionの cumurative distribution function で確率を予測します。
$$\frac{1}{1+\exp^-(\beta_0+\beta_1nwifeinc_i+\beta_2educ_i+\beta_3exper_i+\beta_4exper^2_i+\beta_5age_i+\beta_6kidslt6_i+\beta_7kidsge6_i)}$$

この際のパラメータの推定方法はMaximum Likelihood Estimation です。
logistic 分布の累積確率分布や次のprobit model で使う 正規分布の累積確率分布は0~1の範囲でしか動かないという特徴があります。

summary(logitres<-glm(inlf~nwifeinc+educ+exper+I(exper^2)+age+kidslt6+kidsge6,
                                family=binomial(link=logit),data=mroz))
Call:
glm(formula = inlf ~ nwifeinc + educ + exper + I(exper^2) + age + 
    kidslt6 + kidsge6, family = binomial(link = logit), data = mroz)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-2.1770  -0.9063   0.4473   0.8561   2.4032  

Coefficients:
             Estimate Std. Error z value Pr(>|z|)    
(Intercept)  0.425452   0.860365   0.495  0.62095    
nwifeinc    -0.021345   0.008421  -2.535  0.01126 *  
educ         0.221170   0.043439   5.091 3.55e-07 ***
exper        0.205870   0.032057   6.422 1.34e-10 ***
I(exper^2)  -0.003154   0.001016  -3.104  0.00191 ** 
age         -0.088024   0.014573  -6.040 1.54e-09 ***
kidslt6     -1.443354   0.203583  -7.090 1.34e-12 ***
kidsge6      0.060112   0.074789   0.804  0.42154    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 1029.75  on 752  degrees of freedom
Residual deviance:  803.53  on 745  degrees of freedom
AIC: 819.53

Number of Fisher Scoring iterations: 4

Log likelihood value の算出

logLik(logitres) 
'log Lik.' -401.7652 (df=8)

McFadden's pseudo R-squared

1 - logitres$deviance/logitres$null.deviance
[1] 0.2196814

prediction(just same extreme women as LPM)

predict(logitres, xpred,type = "response")
1           2 
0.005218002 0.950049117 

ちゃんと極端な説明変数の値でも、LPMと違って0~1の間に収まっていることが確認できる。

Probit model

$$\Phi(\beta_0+\beta_1nwifeinc_i+\beta_2educ_i+\beta_3exper_i+\beta_4exper^2_i+\beta_5age_i+\beta_6kidslt6_i+\beta_7kidsge6_i)$$

logit model は働く確率をlogistic 分布の累積確率関数で予測しましたが、probitは正規分布の累積確率分布で予測します。

summary(probitres<-glm(inlf~nwifeinc+educ+exper+I(exper^2)+age+kidslt6+kidsge6,
                                family=binomial(link=probit),data=mroz))
Call:
glm(formula = inlf ~ nwifeinc + educ + exper + I(exper^2) + age + 
    kidslt6 + kidsge6, family = binomial(link = probit), data = mroz)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-2.2156  -0.9151   0.4315   0.8653   2.4553  

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  0.2700736  0.5080782   0.532  0.59503    
nwifeinc    -0.0120236  0.0049392  -2.434  0.01492 *  
educ         0.1309040  0.0253987   5.154 2.55e-07 ***
exper        0.1233472  0.0187587   6.575 4.85e-11 ***
I(exper^2)  -0.0018871  0.0005999  -3.145  0.00166 ** 
age         -0.0528524  0.0084624  -6.246 4.22e-10 ***
kidslt6     -0.8683247  0.1183773  -7.335 2.21e-13 ***
kidsge6      0.0360056  0.0440303   0.818  0.41350    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 1029.7  on 752  degrees of freedom
Residual deviance:  802.6  on 745  degrees of freedom
AIC: 818.6

Number of Fisher Scoring iterations: 4

Log likelihood value

logLik(probitres) 
'log Lik.' -401.3022 (df=8)

McFadden's pseudo R-squared

1 - probitres$deviance/probitres$null.deviance
[1] 0.2205805

同じく働く確率が0~1の間に収まっているかの検証。

predict(probitres,xpred,type = "response")
 1           2 
0.001065043 0.959869044 

大丈夫でした。LPMの弱点は克服できています。

Likelihood Ratio Test for probit model

exper and age are irrelevant 説の検証

$$ H_0:\beta_3=\beta_4=\beta_5=0\ \ \ \ \ H_1:at\ least\ one\ of \ them \ \neq0$$

restr <- glm(inlf~nwifeinc+educ+ kidslt6+kidsge6, 
                          family=binomial(link=logit),data=mroz)
lrtest(restr,probitres)
Likelihood ratio test

Model 1: inlf ~ nwifeinc + educ + kidslt6 + kidsge6
Model 2: inlf ~ nwifeinc + educ + exper + I(exper^2) + age + kidslt6 + 
    kidsge6
  #Df  LogLik Df  Chisq Pr(>Chisq)    
1   5 -464.92                         
2   8 -401.30  3 127.25  < 2.2e-16 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

$H_0$は有意水準0.1%でも棄却できるので、experもageもrelevantでありそう。

2
0
1

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
2
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?