# Interpretation of Economic Significance of Coefficients in Linear Probability Model and Reversing the Dependent Variable

I have the following regression:

open_regime = B1 GovernmentSupport + B2 ln(R&D Budget) + other_controls + FirmFE + TimeFE + ProjectCategoryFE


where open_regime is a dummy variable that takes the value of one if the company has given away the patent rights and zero otherwise. Similarly, GovernmentSupport is a dummy variables that is equivalent to 1 if company gets a support from government and zero otherwise. I also have log transformed R&D budget(for the project) variable and time, firm and project category fixed effects.

I use a linear probability model in order to control for these fixed effects. This is the regression output that I get:

open_regime = 0.007 GovernmentSupport - 0.005 ln(R&D Budget)


In my sample, only 0.03 of the cases firms have chosen to give away patents. I also want to comment on the economic significance of the variables. If there is a 1% price increase in the R&D budget, that to 100 * ((0.03 - 0.00005) - 0.03)/0.03, that is a 0.17% increase in the open regime in the sample. I have three questions:

• Is my interpretation of the ln(R&D Budget) coeffcient is correct?
• What would be the interpretation of GovernmentSupport variable(in 0.08 percent of cases government supports the project)
• If I reverse my dependent variable, say I define a closed_regime variable that takes the value of one if the company has not given away the patent rights and zero otherwise, I observe that my coefficients and standard errors remain the same but they only change their sign... How does my interpretation of economic significance change compared to open regime? (How it is calculated in this case?)