Linear-Log probit and linear probability models

[Katie McGregor]

Hi everyone. I am doing my thesis at the moment and I don't know how to interpret a key variable.

The linear probability model is:

Unemployed_it= B0 +B1log(household income)_it + u_it, where unemployed is a dummy variable =1 if someone is unemployed.

I got -0.0456 as my estimate for B1. How do I interpret this? I also used a probit model and then the margins, dydx(*) command to estimate the above equation. How would I interpret this estimate of B1?

Thanks !

[Fei Wang]

Katie, it means that, a one percent increase in household income is, on average, associated with a probability decrease of 0.000456 in unemployment (or 0.0456 percentage points). "margins, dydx(*)" gives you the same interpretation.