I have a basic question of forecasting in predictive project. I have to predict wine production in several places. The data that I have is climatic inputs like temperature, humidity... This data is been colected by climatic stations located in diferent places near plantations. The point here is that the climatic information is colected by climatic station but the forecasting (production) is colected by a bigger area (where there are some climatics stations on it).

Therefore, climatic stations open and close throughout time. In some cases it is posible that while there is only one climatic station in one area in a specific year, next years there could be more that one climatic station but with the same plantation extension.

So the main points here is that I have climatic information for each station but aggregated information of forecasting (objetive variable).

My question is related to approach. I think that I could aggregate general information about climatic conditions for each area where I have forecasting. The problem here is that i "lose" many information. Another point could be triying to impute productions for each climatic stations and the use all this data to predict production. But is it really a good idea to do this second thing? What are all the aproaches posibilities for this kind of problem?

Any help would be apreciated


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.