内蒙古自治区GDP组合预测分析

GDP Combination Forecast of Inner Mongolia Autonomous Region

GDP是衡量国家和地区经济的重要指标,利用恰当的方法对GDP进行建模预测及分析民族地区GDP与其他指标之间的动态关系对于推动民族区域经济发展具有重大意义。本文选取内蒙古2005年第一季度至2020年第三季度的季度GDP、建筑业(ARCHI)季度总产值和农林牧渔业(FARM)季度总产值数据建立VAR模型,使用格兰杰检验、脉冲响应等方法分析三个指标间的相关关系。为了精确预测内蒙古GDP,本文利用逐步向前模型平均准则对疏系数VAR模型进行组合预测。通过与单序列季节ARIMA模型的预测效果进行比较,发现多数情况下组合VAR模型预测精度更高。由于这两类方法各有优势,本文使用基于IOWA算子的组合预测方法对这两类模型再次组合,实证分析表明,该方法的短期预测效果最优。

GDP is an important indicator to measure the national and regional economy.It is of great significance to use appropriate methods to predict GDP and analyze the dynamic relationship between GDP and other indicators to promote the economic development of ethnic regions.This paper focuses on the quarterly GDP of Inner Mongolia.Based on the VAR model,we first investigate the relation between GDP,the gross output of construction industry and farming by Granger test and impulse response analysis.In order to accurately forecast the GDP of Inner Mongolia,the stepwise forward model averaging method is used to forecast the sparse VAR model.Compared with the seasonal ARIMA model of each single series,it is found that the combined VAR model has higher prediction accuracy in most cases.Because these two methods have their own advantages,the combination forecasting method based on IOWA operator is used to combine these two models again.The empirical analysis shows that the combining short-term forecast obtained by the IOWA operator performs the best.