灰色模型在肺癌死亡率预测中的应用

Application of grey model in forecasting incidences of lung cancer

摘要:

目的：运用灰色模型（Ｇrey model，GM）预测自贡市肺癌死亡率，探索数据波动较大的情况下，适用的GM预测方法。方法：根据自贡市1985-2009年肺结核发病率数据分别建立常规GM[Conventional GM（1，1）]，动态GM[Dynamic GM（1，1）]和新陈代谢GM[Metabolizing GM（1，1）]，选择合适肺癌死亡率的预测模型并进行外推预测。结果：检验性预测中，常规GM（1，1）预测精度为2级，动态 GM（1，1）和新陈代谢 GM（1，1）预测精度为1级，预测值相对误差依次为 26.00%、5.07%和 4.06%。采用新陈代谢 GM（1，1）对自贡市肺癌死亡率进行外推预测，2011-2014 年自贡市肺癌死亡率分别为 81.31/10 万、95.88/10 万、113.06/10 万、133.32/10 万。结论：针对自贡市肺癌死亡率数据波动的情况，采用新陈代谢 GM（1，1）进行预测是比较适用的方法。

Abstract:

Objective：To apply the grey model（GM） in predicting the incidences of lung cancer in Zigong city and to explore the applicable predictive method with fluctuant data. Methods：The conventional GM（1，1），dynamic GM（1，1） and metabolic GM（1，1） were established respectively based on the incidences of lung cancer in Zigong city from 1985 to 2009. The most applicable mortality forecasting model was applied to do the predication. Results：The results of the prediction showed that the predictive precision of the conventional GM（1，1） was the second grade and those of the metabolic GM（1，1） and dynamic GM（1，1） were the first grade. The relative error of the predictive value in three models was 26.00%，5.07% and 4.06%. The metabolic GM（1，1） was applied to predict the mortality rate of lung cancer from 2011 to 2014 in Zigong city and the results were 81.31/100 000，95.88/100 000，113.06/100 000 and 133.32/100 000 respectively. Conclusion: The data of mortality in Zigong city is fluctuant，therefore，metabolic GM（1，1） is an applicable predicting method.