Study on variation in wet and low water of precipitation prediction based on Markov with weights theory

The Markov prediction has been widely applied to simulate natural disasters in recent years as one of the modern prediction theories. This paper first uses Markov chain prediction to study the variation in wet and low water of annual precipitation. The prediction model is set up based on the rainfall data of Xianyang city from 1959 to 2006 and finally, the meteorological drought state of XIY city in 2007 and 2008 is analyzed. The results indicate that the prediction model has more satisfactory fitting accuracy, which could be applied in practical production activities.