论文部分内容阅读
目的了解用GM(1,1)模型建立预测模型时适宜的样本数量。方法用湖北省1991-2011年甲肝、梅毒、乙脑的发病率,分别连续抽取4年、10年的数据建立模型并进行外推预测,对建模情况和预测准确性进行分析比较。结果不同样本数量比较,甲肝、梅毒拟合成功率差异无统计学意义(P>0.05),乙脑差异有统计学意义(P<0.05),预测绝对百分比误差和预测成功率3种疾病差异均无统计学意义(P>0.05)。结论原始数据呈规则的指数变化时,用尽可能小的样本数量为宜,原始数据的指数变化特征出现波动时,应收集足够的样本数量,至能充分表达出指数变化特征为止。
Objective To understand the appropriate sample size for establishing a predictive model using the GM (1,1) model. Methods According to the incidence of hepatitis A, syphilis and Japanese encephalitis in Hubei Province from 1991 to 2011, the data were extracted continuously for 4 years and 10 years respectively. The model was established and extrapolated to predict the accuracy of the model and prediction. Results There was no significant difference in the success rate of hepatitis A and syphilis between different groups (P> 0.05), but there was a significant difference between the two groups (P <0.05). The differences of absolute percentage error and predicted success rate No statistical significance (P> 0.05). Conclusion When the exponential change of the original data is regular, it is appropriate to use as few samples as possible. When the exponential variation of the original data fluctuates, a sufficient number of samples should be collected until the exponential characteristics can be fully expressed.