Using GM(1,1) with sample standard deviation to forecast downtrend rainfall for small sample in Khon Kaen, Thailand

Abstract

The objective of this research is to forecast the rainfall, Khon Kaen cumulative annual rainfall data, using GM(1,1) and Markov chain for small sample size based on sample standard deviation. We compare the accuracy of prediction models between this approach–sdGM(1,1) and standard GM(1,1) in terms of MSE and PARE. We find that sdGM(1,1) is better than GM(1,1).