Linear probability model with fixed effects?

Question

If I want to estimate a linear probability model with (region) fixed effects, is that the same as just running a fixed effects regression? Maybe I'm getting tripped up with the language. My goal is to estimate the effect of a baby bonus. My dependent variable is a binary indicator for NEWBORN and my main independent variable of interest is an indicator for receiving the baby bonus. I control for age, age squared, education, marital status, and household income.

Should I be using:

## 1.) Linear Probability    
LPM <- lm(newborn ~ treatment + age + age_sq + highest_education + marital_stat + 
            hh_income_log, data=fertility_15_45)

or

## 2.) FE Model      
FE_model <- plm(newborn ~ treatment + age + age_sq + highest_education + marital_stat + 
                  hh_income_log, data = fertility_15_45, index="region", model="within")

This doesn't appear to be a specific programming question that's appropriate for Stack Overflow. If you seek recommendations for statistical methods, then you should ask such questions over at Cross Validated instead. You are more likely to get better answers there. — MrFlick
Has been posted here as well: stats.stackexchange.com/questions/462437/… — Helix123

jay.sf jay.sf · Accepted Answer · 2020-04-24T06:34:57

You may want to add a region dummy in your LPM to get region fixed effects. Example:

library(plm)
data(Cigar)

summary(plm(I(sales > 121.2) ~ price + pop, data=Cigar, model="within", index="state"))$coe
#            Estimate   Std. Error     t-value     Pr(>|t|)
# price -2.880255e-03 2.626505e-04 -10.9661107 7.519348e-27
# pop   -6.922327e-06 1.311006e-05  -0.5280165 5.975758e-01

summary(lm(I(sales > 121.2) ~ 0 + price + pop + factor(state), data=Cigar))$coe[1:2, ]
#            Estimate   Std. Error     t-value     Pr(>|t|)
# price -2.880255e-03 2.626505e-04 -10.9661107 7.519348e-27
# pop   -6.922327e-06 1.311006e-05  -0.5280165 5.975758e-01

Linear probability model with fixed effects?

1 Answers