白河烟区烤烟产量多元线性回归模型年景预测

Tobacco Yield Prediction Model Based on SPSS Statistical Software in Baihe County

将烤烟实际产量分离为趋势产量和气象产量，根据1997—2011年白河县气象资料和烤烟产量资料，采用SPSS统计软件分别建立趋势产量和气候产量回归模型，最终建立产量回归模型：Y=-89321．903＋43．455T＋41．972x1＋48．276x2。并对历年产量进行检验，结果表明，预测精度最高为100％，最低为92％，平均精度为97％。该模型具有较高的信度和实用性，可作为白河烤烟产量预报的有效工具之一。

Actual production was separated into the trend of flue - cured tobacco yield and meteorological yield, according to the meteorological data and flue -cured tobacco production in 1997 -2011 in Baihe, SPSS statistical software were used to establish trends of yield and meteorological yield regression model, and the production regression model eventual was es- tablished. Then the production of calendar year was tested, and the maximum prediction accuracy was 100%, the smallest accuracy was 92%, the average accuracy was 97%. The prediction model had a high reliability and practicality, it could be an effective tool for quantitative prediction.