Request for Recommendations on Forecasting Count Data

[Liu499-create]

Dear Author,

I hope this message finds you well. I’ve been reading your book and found it incredibly insightful. However, I have a follow-up question regarding a topic that is briefly mentioned in your text.

In the book, you mention that the methods discussed assume data with a continuous sample space, while some data, such as counts (e.g., the number of customers entering a store per day), inherently belong to a discrete, non-negative integer space. However, if our data contains small counts
(0,1,2,…), then we need to use forecasting methods that are more appropriate for a sample space of non-negative integers.The book also states that forecasting methods suitable for count data are beyond its scope.

Could you kindly recommend any resources, books, or R packages that specialize in forecasting count data, or other appropriate methods for time series forecasting with integer values?

I would greatly appreciate any suggestions you might have on this topic.

Thank you for your time and assistance.

Best regards,

[robjhyndman]

Papers:

• Christou, V, and K Fokianos. 2015. “On Count Time Series Prediction.” Journal of Statistical Computation and Simulation 85 (2): 357–73. https://doi.org/10.1080/00949655.2013.823612.
• Hudecová, Š. (2025). Structure Assessment in Count Time Series. Annual Review of Statistics and Its Application, 13. https://doi.org/10.1146/annurev-statistics-042424-114518

R packages:

• https://cran.r-project.org/web/packages/tscount/
• https://cran.r-project.org/web/packages/fableCount/