Consider the hypothesis that poverty is a function of race and sex. Sample data on the subject are collected and coded using the three dummy variables P, R, and S.

The P dummy represents poverty (P = 1 for poverty), the R dummy represents race (R = 1 for black), and the S dummy represents sex (S = 1 for female).

Suppose the following multiple regression equation is estimated:

(a) Interpret the coefficients of R and S.

(b) Is there any problem associated with this interpretation of the equation?
In other words, is there a violation of the assumptions of linear probability models?

Transcribed Image Text:

P = .05+23R + .45S