Forecasting Chinese GDP with Mixed Frequency Data Set:A Generalized Lasso Granger Method

摘要：

　　In this paper, we introduce an effective machine learning method which can capture the temporal causal structures between irregu lar time series to forecast China GDP growth rate with Mixed Frequency data set.The introduced method first generalized the inner product op erator via kernels so that regression-based temporal casual models can be applicable to irregular time series, then the temporal casual relation ships among the irregular time series are studied by Generalized Lasso Granger (GLG) graphical models.The main advantage of this approach is that it does not directly estimate the values of missing data of low frequency time series or has restricted assumptions about the generation process of the time series.By applying this method to a 17 macroe conomic indicators GLG model, the forecasting accuracy is better than the autoregressive (AR) benchmark model and a widely used mixed-data sampling (MIDAS) model.

关键词： Forecast GDP growth Mixed frequency data Generalized Lasso Granger

作者: Zhe Gao Jianjun Yang Shaohua Tan

作者单位: Department of Intelligence Science, Center for Information Science, Room 2314,Science Building 2, Peking University, Beijing 100871, China

会议类型: 国际会议

会议名称: 4th international Conference,ICSI2013(第4届群体智能国际会议)

会议地点: 哈尔滨

会议语种:英文

页码: 163-172

在线出版日期: 2013-06-12（万方平台首次上网日期，不代表论文的发表时间）