Time Series Analysis Applying In The Mastacarcinoma Hospitalized Patients' Medical Costs

Objective Aim at the influence factors of the mastacarcinoma patients' hospitalized costs, the more reasonable cost time series model will be investigated and established. Methods Applying linear regress analysis and time series analysis (including: polynomial curve model,exponential smooth model,gray model,Markov model and state-space model), 350 mastacarcinoma hospitalized patients' medical costs were be studied. Results 1,The results of the linear regress analysis are badï¼›otherwise, hospitalization days, cancer transfer condition and treatment measure arethree main factors of the mastacarcinoma hospitalized patients' medical costs; 2,The polynomial curve model's results is good but the model can not predict well; the exponential smooth model' results is better and the model may reasonably predict in future; the results of the gray model is very bad for the cost time series existing the certainly fluctuation; the Markov model can more ideally predict the cost zone range; both the establishment and the forecast of the state-space model are perfectly; 3,Comparing the linear regress analysis with the time series analysis, the latter's MAPE are less than the former. Conclusions 1,Whether establishment effects or short-term predict, all of the time series analysis ways are more accurate and perfect than the linear regress analysis. Therefore the time series analysis should be tried to apply in the fact works ; 2,The pre-treatment of the cost data has the certain possibility.