Application of Weighted Markov Chain in Precipitation Forecast in Beijing

Abstract

Aiming at the complexity and randomness of the precipitation sequence in Beijing, a Markov chain prediction model was established. Based on the precipitation in Beijing from 1951 to 2013, the precipitation in 2014 and 2015 was predicted. The results show that: (1) The sequence of precipitation in Beijing from 1951 to 2013 satisfies the Markov property when the significance level a=0.1; (2) It is predicted that the precipitation states in Beijing in 2014 and 2015 are weak dry, with the precipitation of 537.9mm and 525.5mm, respectively. The relative errors with actual precipitation are 16.55% and 14.59%, respectively. It can be seen that it is reasonable to use the weighted Markov chain model to predict the precipitation in Beijing.