I have an hourly time series that contains 2 forms of seasonality: hourly and weekly. In other words, the hour of the day and the day of the week both have a large impact on the time series values.

Is there a forecasting model (hopefully a pre-built one in R) that can handle a time series with multiple seasonalities?

One R function I have tried was auto.arima with xreg being a day-of-week indicator matrix.


We have a tag , you may want to browse through it.

Standard models include and . Take a look at their tag wikis for more information and pointers to literature.

| cite | improve this answer | |

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.