Search

This article examines whether combining forecasts of real GDP from different models can improve forecast accuracy and considers which model-combination methods provide the best performance. In line with previous literature, the authors find that combining forecasts generally improves forecast accuracy relative to various benchmarks. Unlike several previous studies, however, they find that, rather than assigning equal weights to each model, unequal weighting based on the past forecast performance of models tends to improve accuracy when forecasts across models are substantially different.

The formulation of monetary policy at the Bank of Canada relies on the analysis of a broad set of economic information. Greater availability of immediate and detailed information would improve real-time economic decision making. Technological advances have provided an opportunity to exploit “big data” - the vast amount of digital data from business transactions, social media and networked computers. Big data can be a complement to traditional information sources, offering fresh insight for the monitoring of economic activity and inflation.

Interest rates in China are composed of a mix of both market-determined interest rates (interbank rates and bond yields), and regulated interest rates (retail lending and deposit rates), reflecting China’s gradual process of interest rate liberalization.

We develop a finite-sample procedure to test for mean-variance efficiency and spanning without imposing any parametric assumptions on the distribution of model disturbances.

Forecasts of the quarterly real price of oil are routinely used by international organizations and central banks worldwide in assessing the global and domestic economic outlook, yet little is known about how best to generate such forecasts. Our analysis breaks new ground in several dimensions.

The authors use the Financial Stress Index created by the International Monetary Fund to predict the likelihood of financial stress events for five developed countries: Canada, France, Germany, the United Kingdom and the United States.

A longstanding finding in the forecasting literature is that averaging forecasts from different models often improves upon forecasts based on a single model, with equal weight averaging working particularly well. This paper analyzes the effects of trimming the set of models prior to averaging.

This paper proposes a novel regression-based approach to the estimation of Gaussian dynamic term structure models that avoids numerical optimization.

Default rates are series commonly used in stress testing. In Canada, as in many other countries, there are no historical series available for sectoral default rates on bank loans to firms.

Using a unique micro-dataset containing real and financial information on Canadian households for 2000–07, the authors address two questions: (1) What is the proportion of households whose consumption displays excess sensitivity to income, and who are likely liquidity constrained?