Bootstrapping multivariate spectra
Article Abstract:
The Franke-Hardle (1992) technique for bootstrap inference on estimated univariate spectral-density functions was generalized for multivariate spectra. The generalization is nontrivial and makes use of the Franke-Hardle bootstrap in frequency-domain econometric cases. Empirical results from a small Monte Carlo experiment showed that the performance of the multivariate Franke-Hardle bootstrap method is better than conventional, first-order asymptotics for sample sizes relevant in econometrics.
Publication Name: Review of Economics and Statistics
Subject: Mathematics
ISSN: 0034-6535
Year: 1998
User Contributions:
Comment about this article or add new information about this topic:
Modeling nonlinearity of business cycles: choosing between the CDR and STAR models
Article Abstract:
The current depth of the recession (CDR) model is compared with the smooth transition autoregression (STAR) model in simulating US real GNP and industrial production. Results indicate that the CDR model fits better than STAR and produces more accurate forecasts. The STAR model also has no support in the sample period. Both models give different forecasting performance and dynamics but economists who are interested in studying the nonlinear behavior of output should use the CDR model.
Publication Name: Review of Economics and Statistics
Subject: Mathematics
ISSN: 0034-6535
Year: 1999
User Contributions:
Comment about this article or add new information about this topic:
Measuring business cycles: a modern perspective
Article Abstract:
Business cycle data is measured and analyzed using a linear model that incorporates regime switching and factor structure. Factor structures refer to comovements in the economic setup due to coordinated activity among various economic variables. On the other hand, regime switching deals with fluctuations in economic behavior during periods of growth and recession. The model is compatible with accepted macroeconomic theory and facts.
Publication Name: Review of Economics and Statistics
Subject: Mathematics
ISSN: 0034-6535
Year: 1996
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: Unemployment variation over the business cycles: a comparison of forecasting models
- Abstracts: Forecasting stock market volatility using (non-linear) Garch models. Do seasonal unit roots matter for forecasting monthly industrial production?
- Abstracts: Modelling multivariate cointegrated systems: insights from non-linear dynamics
- Abstracts: Consistent specification testing via nonparametric series regression. Consistent testing for serial correlation of unknown form
- Abstracts: Estimating deterministic trends in the presence of serially correlated errors. Money, prices, interest rates and the business cycle