I have read a few papers which claim that PV and wind farm forecasted errors show negative correlation if forecast is less than day ahead and aggregated output of different PV and WF zones are considered. I am working on GEFCom 2014 data. Suggest me how to develop working relationship between PV and WF.