The advantage of using the probit model versus the linear probability model for estimating regressions for binary dependent variables is:
A. probit model yields more precise estimates (i.e. lower standard errors of the estimated coefficients)
B. regression coefficients in the probit model are easier to interpret
C. probit model yields predicted probabilities that lie between 0 and 1
D. all of the above
Option C: probit model yields predicted probabilities that lie between 0 and 1
Linear probability model is the easiest to interpret. Cumulative distribution function of the linear probability equation to condense the estimated values between 0 and 1 is the probit model. So since there is no new predictive model or equation there is no question of standard errors or precise estimates. Linear Probability Model model can yield probability estimates greater than 1 or less than 0. So in spite of being easy to interpret it is much less accurate as number of data points are wrongly predicted. Thus using the cumulative distribution function of the linear probability equation all the predicted probabilities are confined between 0 and 1 as desired.
Get Answers For Free
Most questions answered within 1 hours.