摘要
文章以内蒙古林业产业总产值为研究对象,依据1994—2015年《中国林业统计年鉴》数据,建立了基于马尔科夫修正的三次指数平滑预测模型。一方面基于1994—2013年数据,利用三次指数平滑法预测了2014—2017年林业产业总产值;另一方面由于林业产业的发展具有随机性、动态性等特点,采用马尔科夫链对预测结果进行修正,与2014年和2015年实际林业产业总产值相比,其相对误差分别减至4.86%和2.43%,且模型平均绝对百分误差为8.33%,达到精度较高的预测效果。表明基于马尔科夫修正的三次指数平滑预测模型在林业产业总产值的预测中具有可行性和有效性。
Taking the total output value of forestry industry in Inner Mongolia as the research object, based on the data of China Forestry Statistics Yearbook from 1994 to 2015, this paper established the three order exponential smoothing prediction model based on Markov Correction. On one hand,based on the data from 1994 to 2013, three order exponential smoothing method was used to predict the total output value of forestry industry from 2014 to 2017. On the other hand, due to the randomness and dynamic characteristics of the forestry industry, the Mar- kov Chain was used to modify the forecast result. Compared with the actual output value of forestry industry in 2014 and 2015, the relative error was respectively reduced to 4.86% and 2.43%, and the average absolute percentage error of the model was 8.33%, which achieved the higher precision prediction effect. It showed that the three order exponential smoothing prediction model based on Markov correction had been feasible and effective in predicting the total output value of forestry industry.
作者
张建成
智荣
ZhangJiancheng, Zhi Rong(School of Economics and Management, Inner Mongolia Agricultural University, Huhhot 01001)
出处
《林业经济》
北大核心
2018年第3期35-41,共7页
Forestry Economics
基金
国家社会科学基金西部项目“基于对农民林业经营行为激励视角的集体林权改革效果优化策略研究”(编号:16XJY010)
关键词
林业总产值
三次指数平滑法
马尔科夫链
total output value of forestry industry
three order exponential smoothing method
Markov Chain