The Information Content of Financial Aggregates in Australia
Ellis W. Tallman and Naveen Chandra
Federal Reserve Bank of Atlanta
Working Paper 96-14
This paper examines whether financial aggregates provide information useful for predicting the subsequent behavior of real output and inflation. We employ vector autoregression (VAR) techniques to summarize the information in the data, providing evidence on the incremental forecasting value of financial aggregates for forecasting real output and inflation. The in-sample results suggest that there are only a few situations in which knowledge of the aggregates helps forecast real output and inflation. We then test the forecast performance of the VAR systems for two years out-of-sample in order to mimic more closely the real-time forecasting problem faced by policymakers. We compare the out-of-sample forecast accuracy of VAR systems including a financial aggregate with the corresponding system excluding the financial aggregate. Overall, both in-sample and out-of-sample results suggest no robust finding of exploitable information that is useful for policymakers in any of the financial aggregates under examination.
JEL classification: E40, E44, E51