I'm developing a probit model to predict the partial effect of independent variables on the expected probability of my dependent variable being one (that is the event that a strong growth in GDP occurs). Does it make sense to regress the binary dependent variable on an average growth rate of an independent variable, $g_{t-k,t}$, the average growth rate of import of computers from $t-k $ to $t$? But, in this case the change in the independent variable, $\Delta g_{t-k,t}$ is an acceleration. Does this make sense in your opinion?

  • 1
    $\begingroup$ I'm not sure this is a statistics question. It might be better in the Economics stack exchange. $\endgroup$
    – Peter Flom
    Commented Sep 9, 2023 at 11:27
  • 2
    $\begingroup$ I’m voting to close this question because it belongs on the Economics site $\endgroup$
    – Peter Flom
    Commented Jan 21 at 0:20