Vincent Labhard

Forecasting using Bayesian and information theoretic model averaging: an application to UK inflation

In recent years there has been increasing interest in forecasting methods that utilise large data sets, driven partly by the recognition that policymaking institutions need to process large quantities of information. Factor analysis is a popular way of doing this. Forecast combination is another, and it is on this that we concentrate. Bayesian model averaging methods have been widely employed in this area, but a neglected alternative approach employed in this paper uses information theoretic based weights. We consider the use of model averaging in forecasting UK inflation with a large data set from this perspective. We find that an information theoretic model averaging scheme can be a powerful alternative both to the more widely used Bayesian model averaging scheme and to factor models.

A state space approach to extracting the signal from uncertain data

Most macroeconomic data are uncertain—they are estimates rather than perfect measures of underlying economic variables. One symptom of that uncertainty is the propensity of statistical agencies to revise their estimates in the light of new information or methodological advances. This paper sets out an approach for extracting the signal from uncertain data. It describes a two-step estimation procedure in which the history of past revisions is first used to estimate the parameters of a measurement equation describing the official published estimates. These parameters are then imposed in a maximum likelihood estimation of a state space model for the macroeconomic variable.

Wealth and Consumption: An Assessment of the International Evidence

The main objective of this paper is to offer a critique of the existing literature on the link between wealth and consumption, as captured by the long-run marginal propensity to consume from financial wealth (mpcw). The international evidence suggests that the mpcw varies considerably across countries, and new estimates are presented, based on structural vector autoregressions (VARs) for eleven OECD countries, which tend to confirm this finding. It is argued that there is little theoretical rationale for a wide cross-country dispersion of the mpcw, and that the cross-country differences in empirical estimates may in fact reflect difficulties in the measurement of wealth across countries and a failure to account for the shocks causing changes in both consumption and wealth. Using a suitable panel technique, it is found that the hypothesis of a common long-run mpcw across countries cannot be rejected consistently, and a plausible estimate is obtained for the cross-section of eleven OECD countries. This estimate is a little over 6%, broadly consistent with estimates used in a wide range of policy models.

Forecasting Using Bayesian and Information-Theoretic Model Averaging

Model averaging often improves forecast accuracy over individual forecasts. It may also be seen as a means of forecasting in data-rich environments. Bayesian model averaging methods have been widely advocated, but a neglected frequentist approach is to use information-theoretic-based weights. We consider the use of information-theoretic model averaging in forecasting U.K. inflation, with a large dataset, and find that it can be a powerful alternative to Bayesian averaging schemes.

What Caused the 2000/01 Slowdown? Results from a VAR Analysis of G7 GDP Components

In this paper a VAR-based analysis of shocks to G7 GDP components during the 2000/01 slowdown is presented. The patterns of shocks across the components and across the G7 countries are documented, and measures provided of their persistence. The shocks during the preceding expansion are also considered, and are used to discuss possible business cycle asymmetries, and a comparison made with the pattern of shocks during the previous slowdown in 1990. The analysis is then extended to derive shocks to components that explicitly take into account the roles played by monetary policy and oil prices in 2000/01.