I want to describe the direction of some stock returns, using as predictors several independent variables which are uncorrelated. The relation in which I am interested is between the stock returns and an independent variable which is autocorrelated. To describe the direction I use a binary logistic regression, with a binary independent variable that takes values: 0 if the stock return is negative and 1 if the stock return is positive. My question is: what I obtain if I bootstrap a binary logistic regression, in that situation? It could be correct or it does not make any sense? And why?

  • $\begingroup$ What do you mean by boostrapping the logistic regression in this context? $\endgroup$ – Ric Mar 9 '16 at 16:18
  • $\begingroup$ I was wondering if it makes sense to use bootstrap if I have time series, and if it could be useful to solve the problem of the autocorrelated independent variable. Moreover, my real doubt is about what is the output of this estimation. $\endgroup$ – Steph Mar 9 '16 at 17:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.