The Application Of The Method Of Time Series Analysis In The Study Of Chongqing Temperatures

With further study global climate change, people have come to realize the impact that extreme weather events brought people's daily production and life, socio-economic and ecological environment brought on have exceeded the impact that the long-term climate change brought human life, among all the influence extreme weather events brought in, it is often due to extreme heat or extreme cold caused by the occurrence, therefore, the present study of extreme temperatures is looming, domestic and foreign experts pay much more emphasis on the study of extreme temperatures.This article analyzes the variation of the monthly average maximum and minimum temperatures in Chongqing during 2011-2015 in the case of global warming, and found out the model that meets the change of the monthly average highest and lowest temperature under based on this variation, to forecast the future by the optimal model. The main research method in this article is the seasonal effect analysis and stochastic analysis in time series analysis methods, making trend analysis on monthly average maximum and minimum temperatures during the past five years in Chongqing by stochastic analysis. In this thesis, we have mainly studied the model of ARMA(p q) what the differential operation of the monthly average maximum and minimum temperature data sequence meets, and make the model significance test and simulation results test and prediction residuals test. Under the guidelines of the maximum of AIC, we adopt the optimal model among all models that have pass the significance test, using the selected model to predict the final choice and make a prediction residual test, and study the accuracy of the model in prediction at last. Finally, because of lower prediction accuracy of the fitted model, we use seasonal index to make predictive analysis to the future development of monthly average maximum and minimum temperatures by the method of seasonal effects analysis. Finally, we can find a fitting model to mostly simulate the change trend of Chongqing monthly average minimum, maximum temperature sequence, and predict the future of Chongqing monthly average minimum, maximum temperature by seasonal index with highly accurate, as long as a given month average maximum or minimum temperature, we can take advantage of the seasonal index to predict average maximum or minimum temperature in another month.