目的 :建立山地进攻战役减员的统计学模型 ,提供计算机模拟研究的基础。方法 :收集了我军某次战争某战役方向战役减员的经验数据 ,计算每日减员率 ,采用柯尔莫哥洛夫_斯米尔诺夫方法进行统计分布的拟合优度检验 ,运用博克斯_詹金斯方法建立时间序列模型。结果 :战役减员率分布符合伽玛分布 ,时间序列模型为ARIMA( 0 ,1,1)。结论 :时间序列模型能够较好地说明减员分布的统计学特征 OBJECTIVE: To establish a statistical model of attrition in offensive battles in mountainous areas and provide the basis for computer simulation research. Methods: The empirical data of attrition reduction in a certain battle direction of a certain war in our army was collected, the daily reduction rate was calculated, the goodness-of-fit test of statistical distribution was conducted by Kolmogorov-Smirnov method, Jenkins method to establish a time series model. Results: Campaign attribution rate distribution is in accordance with Gamma distribution. The time series model is ARIMA (0,1,1). Conclusion: The time series model can better explain the statistical characteristics of downsizing distribution