中文摘要：

目的本次研究旨在通过建立时间序列相关模型对天津市卫生资源的人、财、物三方面进行趋势分析和预测。方法收集了天津市1980～2008年医院、卫生院卫生技术人员数、床位数、卫生事业费用，运用时间序列相关模型分别对其进行了预测。结果对数据进行了时间序列相关分析模型的筛选后，最后选用ARIMA（1，1，1）模型对天津市医院、卫生院卫生技术人员数进行趋势分析和预测，采用随机游走模型对天津市医院、卫生院床位数、卫生事业费用进行趋势分析和预测，其样本内预测其平均相对误差分别为0．023、0．022、0．074，卫生技术人员数和床位数样本外预测结果平均相对误差分别为0．0053、0．012，卫生事业费用数样本外预测相对误差为0．144。结论时间序列方法在天津市卫生资源人、财、物方面预测中显示了较好的精度，预测结果可为我市“十二五”期间卫生资源的配置提供科学的参考。

英文摘要：

Objective The aim of the study was to analyze the tendency and forecast the manpower,material and financial of the health resources by using the time series models. Methods We collected heath manpower of hospital,beds and health expenditure in Tianjin from 1980 to 2008 and used time series model to forecast them. Results After screen- ing the related time series models according the conditions of the data, the sample internal mean relative errors of the ARIMA（ 1,1,1 ） model which forecasts the health manpower was 0. 023. The sample internal mean rela- tive errors of the Random Walk with Drift which forecasts the beds and ex- penditure of health resources were 0. 022,0. 074 respectively. And the sam- ple external mean relative errors of the health manpower and beds were 0. 0053,0.012 respectively, the sample external relative errors of the health expenditure was 0. 144. Conclusion The time series methods achieved good precision of forecast, the results of the forecast could provided refer- ence for the allocation of the health resources during the next National Twelfth Five-Year Plan period in Tianjin.