FRS161_NDVI_MLR

A premature analysis of which 2 non-precipitation variables out of 41 variables yield the highest accuracy in hindcasting NDVI in Enotre, Italy 2019. All MLRs are trained on 2012-2018 data and are asked to predict 2019 NDVI. There exists 17 unfiltered possible MLR variable combinations that yield a 2019 hindcasting error of less than 11%.