18 mai 2011
Inflation Expectations in Turkey
We investigated the rationality of financial and real sectors’ CPI inflation expectations in Turkey using the multivariate panel cointegration method. The use of panel techniques strengthened our empirical results by not only increasing sample size but also allowing heterogeneity across groups of respondents. Having found the expectations irrational in the stricter sense, we proceeded to analyze the significance of both past and future inflation rates as determinants of agents’ future inflation forecasts. Both recursive and rolling estimates show that forecasters’ weight on future/target inflation rates versus past actual and expected inflation rates changes over time as unexpected shocks derail inflation from its disinflationary path. Lastly, we find asymmetry in expectations such that the response of inflation expectations to an increase in the inflation rate is twice the size of the response to a decrease in the inflation rate. This may indicate long delays in restoring credibility of central banks after a positive shock on the inflation rate. JEL Classifications: C23, D84, E31 Keywords: Inflation expectations, Inflation formation, Panel cointegration, Recursive regression
18 mai 2011
Forecasting with Leading Indicators by means of the Principal Covariate Index
A new method of leading index construction is proposed, which explicitly takes into account the purpose of using the index for forecasting a coincident economic indicator. This so-called principal covariate index combines the need for compressing the information in a large number of individual leading indicator variables with the objective of forecasting. In an empirical application to forecast future growth rates of the Conference Board’s Composite Coincident Index and its constituents, the forecasts of the principal covariate index are more accurate than those obtained either from the Composite Leading Index of the Conference Board or from an alternative index-based on principal components. JEL Classification: C32, C53, E27 Keywords: index construction, business cycles, principal component, principal covariate, time series forecasting, variable selection
18 mai 2011
Are Qualitative Inflation Expectations Useful to Predict Inflation?
This paper examines the properties of qualitative inflation expectations collected from economic experts for Germany. It describes their characteristics relating to rationality and Granger causality. An out-of-sample simulation study investigates whether this indicator is suitable for inflation forecasting. Results from other standard forecasting models are considered and compared with models employing survey measures. We find that a model using survey expectations outperforms most of the competing models. Moreover, we find some evidence that the survey indicator already contains information from other model types (e.g. Phillips curve models). However, the forecast quality may be further improved by completely taking into account information from some financial indicators.
18 mai 2011
The Forecasting Performance of Composite Leading Indicators
Using OECD Composite Leading Indicators (CLI), we assess empirically whether the ability of the country-specific CLIs to predict economic activity has diminished in recent years, e.g. due to rapid advances in globalisation. Overall, we find evidence that the CLI encompasses useful information for forecasting industrial production, particularly over horizons of four to eight months ahead. The evidence is particularly strong when taking cointegration relationships into account. At the same time, we find indications that the forecast accuracy has declined over time for several countries. Augmenting the country-specific CLI with a leading indicator of the external environment and employing forecast combination techniques improves the forecast performance for several economies. Over time, the increasing importance of international dependencies is documented by relative performance gains of the extended model for selected countries.
18 mai 2011
Aggregate Comovements, Anticipation, and Business Cycles
This paper shows that negative comovements between major macroeconomic variables at business-cycle frequencies are commonly observed, but that standard Real Business Cycle (RBC) theory fails to predict this feature of the data. We show that allowing for "anticipation effects" in response to "news shocks" enables standard RBC models to predict both the observed patterns of negative comovement and overall positive correlations. Anticipation also improves magnification of shocks in the model without harming predictions for the other second moments central to RBC studies. Anticipation effects improve on standard RBC frameworks by offering an empirically plausible explanation for the nontrivial fraction of time that aggregate variables are observed to comove negatively.