0

Data - Monthly Rainfall of a region for the past 20 years.

Objective - To Forecast for the next 2 years.

In the collected data, 10 out of 13 values for January are 0s and the other three are (1.9, 0.9, 0.2). When done forecasting for the next 2 years, The values obtained for January are negative (i.e. -1.66, -0.49). Since I am dealing with Rainfall data is there a way to get the forecasting values for January to be 0 or positive.

noob
  • 65
  • 6
  • How are you defining 'better' here? Is there some holdout testing you are doing? – Tylerr Nov 04 '21 at 18:32
  • See https://stats.stackexchange.com/questions/162598/a-model-for-non-negative-data-with-many-zeros-pros-and-cons-of-tweedie-glm also https://www.google.com/search?q=time+series+forecasting+negative+values+site:stackoverflow.com&client=ubuntu&hs=QWx&channel=fs&sxsrf=AOaemvI1uBYm7amxCG04XUv0dWRuWRlHag:1636097653178&sa=X&ved=2ahUKEwiesb3b2oD0AhUKIbkGHYW5AVcQrQIoBHoECDYQBQ&biw=1235&bih=645&dpr=1.5 – kjetil b halvorsen Nov 05 '21 at 07:35

0 Answers0