语言随机效应模型_医学统计与R语言：Probit回归模型及边际效应（Marginal effects）...

微信公众号：医学统计与R语言如果你觉得对你有帮助，欢迎转发

A probit regression is a version of the generalized linear model used to model dichotomous outcome variables. It uses the inverse standard normal distribution as a linear combination of the predictors. The binary outcome variable Y is assumed to have a Bernoulli distribution with parameter p (where the success probability is p∈(0,1)p∈(0,1)). Hence, the probit link function is

Remember back to intro stats when you had to look up in Z tables the area under the normal curve for a specific Z value? That area represents a cumulative probability: the probability that Z is less than or equal to the specified Z value.

Coefficients for probit models can be interpreted as the difference in Z score associated with each one-unit difference in the predictor variable.

Although the effect on Z of a change in  X is linear, the link between  z and the dependent variable  Y is nonlinear since  Φ is a nonlinear function of  X.

输入1：

rm(list=ls())
setwd("C:/Users/mooshaa/Desktop")
library(rio)
bilog import("log.sav")
head(bilog)
bilog[,c(1,2,6)] 1,2,6)] ,as.factor)
xtabs(~Gender + Smoking_status, data = bilog)

结果1：

 Smoking_status
Gender  0  1
     0 94 27
     1 17 80

输入2：

logitmodel <- glm(Smoking_status~., data = bilog, family = binomial(link="logit"))probitmodel <- glm(Smoking_status~., data = bilog, family = binomial(link="probit"))summary(probitmodel)

结果2：

Call:
glm(formula = Smoking_status ~ ., family = binomial(link = "probit"), 
    data = bilog)

Deviance Residuals: 
    Min       1Q