内蒙古赤峰市达里湖渔业产量的灰色预测与分析

Prediction of Fishery Production of Dalihu Lake in Chifeng City,Inner Mongolia Using a Grey Prediction Model

由于病害和产卵条件恶化等原因,内蒙古赤峰市达里湖的鱼类资源出现了衰退,经调整产业政策,渔业产量在大幅度下降后又有所回升。本文使用内蒙古赤峰市达里湖2000—2009年的渔业产量数据,并应用灰色预测模型对其未来5 a的渔业产量进行了预测。结果表明,使用全程数据进行预测的渔业产量将持续下降,与最近几年的渔业产量趋势明显不符,这是由于在数据序列的中段发生的渔业产业政策调整造成了产量大幅度下降,其惯性对预测结果产生了决定性影响,而仅使用产业政策调整后的数据进行预测,则渔业产量呈上升趋势,并有望恢复到历史最高水平,说明渔业产业政策调整有利于渔业资源的保护和可持续利用。本文的结果同时表明,应用于灰色预测的数据的外部条件的突然变化(如本文中的渔业产业结构调整),会对预测结果产生很大影响,在选择数据序列时应予以关注。此外如何对数据外部条件的突然变化进行量化评估,并嵌入灰色预测模型,使灰色预测模型能够应对数据外部条件的突然变化,也是今后的1个研究课题。

Because of diseases and deteriorating spawning conditions,the fisheries resources in the Dali Lake in Chifeng city of Inner Mongolia had declined.After the adjustment of the industry policy,there was some recovery in fishery production after a large drop.This study used fishery production data in 2000-2009 in Dali Lake in Chifeng city of Inner Mongolia to predict the future 5 years fishery production using a grey prediction model.The result showed that the fishery production would continue to decline,which contradicted with the trend of the recent years.Because the adjustment of fishery industry policy made a large drop of production in the middle of the data series,the inertia influenced critically on the prediction result.When only using the data after the change of industry policy,the fishery production had a trend of increase which may be back to the highest historical level.These illustrated that the adjustment of fishery industry policy can be in favor of fishery resources conservation and sustainable exploration.At the same time the results also showed that when the outside conditions of the grey prediction data changed suddenly（such as the adjustment of fishery industry policy in this study）,the prediction outcome can be influenced dramatically.Attentions should be paid when selecting the data time series.Additionally the quantitative assessment to the sudden changes of the outside data conditions,and its incorporation into the grey prediction model,is an important research topic in the future.And the same important thing include how to make the grey prediction model to deal with the sudden changes of the outside data conditions.