I am looking at historical hourly electricity demand data over multiple years and trying to regress it with hourly temperature data to understand the true relationship. I am curious if the hour of the day, week, season, year, can be considered as fixed effects?

My dataset is basically a time-series of hourly electricity demand (has timestamps) and hourly temperature (again has timestamps), and I have concatenated them into one large data frame for the ease of applying OLS models.

  • $\begingroup$ Yes. Care to share a small subset of your data? $\endgroup$ – Thomas Bilach Jan 26 at 6:43
  • 2
    $\begingroup$ You have a slight issue that sunrise and sunset vary through the year so the hourly effect is not independent of the season. You may also want to include day of the week in your model. $\endgroup$ – Henry Jan 26 at 15:05

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.