I am building a multivariate statistical model to forecast the trading volume of the S&P 500 stock based on its previous values and on other covariates. Being new to finance, I am having problems selecting the appropriate exogenous regressors. So far, I am thinking of including daily returns to the model, as they tend to correlate with trading volume. Aside from that what could be other (ideally stationary) series that could help improve model accuracy?

  • $\begingroup$ Calendar events and seasonality. $\endgroup$
    – Bob Jansen
    Nov 10, 2021 at 16:07


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.