【主讲人简介】

李富春，博士，加拿大支付系统高级科研顾问（Senior research advisor），主要从事于金融系统风险评估和金融系统稳定性研究。李富春获得南开大学理学学士学位，University of Guelph经济学硕士学位，Western University经济学博士学位。李富春博士已有多篇科研文章发表在顶尖期刊上，如Journal of Econometrics，Econometric Theory，Journal of Financial Econometrics，Journal of Banking and Finance 等。

【英文摘要】

In this paper, we examine the information content of macro-financial variables for forecasting the growth of payments transaction in Canada. Three different models are used to evaluate the predictive accuracy of the macro-financial variables for forecasting the growth of payments transaction, namely, the ARIMA(1, 1, 1) model, a linear regression model with macro-financial variables, a threshold regression model with macro-financial variables. Our empirical results confirm the ability of some key macro-financial variables to forecast future growth of total payment value and total payment volume. Both the linear model and threshold regression model with macro-financial variables perform better than the benchmark model, ARIMA (1, 1, 1) model.

Further, we propose a test statistic to test for whether omitted variables occur in the volatility function in a traditional time series regression model. The test statistic asymptotically converges to a standard normal distribution. A Monte Carlo study shows that the test statistic performs reasonably well.

撰稿：班伟审核：张同斌单位：经济学院