Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It only takes a minute to sign up.
Sign up to join this community
Is there a way to incorporate known values into a time series model?
For example if we look at hotel bookings. It makes sense to build a seasonal forecasting model, but I'd like to try to factor in "early bookings". Say we look 180 days prior to a specific week, we know that a certain # of people have already booked for that week and would be a helpful indicator to what actual bookings would end up being.
Forecasting hotel and/or restaurant bookings using advance information is a lively research field. Searching at Google Scholar for "forecasting hotel booking" turns up a number of relevant papers, mostly in tourism journals. You could vary the search parameters by adding "advance" to weed out papers that do not consider this source of information, and/or substituting "restaurant" or "airline" for "hotel", since similar questions pop up in these industries.
