Forecasts of inflation from VAR models

Article Abstract:

The inaccuracy exhibited by vector autoregression (VAR) models in predicting inflation results from the misspecification which occurs during the fitting of a VAR model's price equation to US post-war economic variables. As evidence, the price equations of two VAR models are presented. Two counter the instability of these price equations, two changes in monetary policy changes are proposed to lessen misspecification and improve forecasting accuracy.

Inflation (Finance), Inflation (Economics)

---

An integrated Bayesian vector autoregression and error correction model for forecasting electricity consumption and prices

Article Abstract:

It was argued that the Bayesian vector autoregression (VAR), error correction model (ECM) and cointegration model are restricted versions of the traditional VAR model. This concept, when used to derive a four-step method for specifying VAR forecasting models, resulted in five, different VAR models. To evaluate their performance, the five were employed to forecast electricity consumption and price using US data between 1989-91.

Bayesian statistical decision theory, Bayesian analysis, Energy consumption

---

Combining VAR estimation and state space model reduction for simple good predictions

Article Abstract:

Monte Carlo methods were employed to heuristically evaluate vector autoregression estimation and state space model reduction techniques. The investigation derived 'simple to use' strategies for obtaining models with good forecasting characteristics. The results revealed that the difficulty in parameter estimation of an accurate model yields more prediction error than that of a 'parsimonious approximate model.'

Usage, Monte Carlo method, Monte Carlo methods, Time-series analysis, Time series analysis, Parameter estimation

Similar abstracts:

• Abstracts: Forecasting with generalized Bayesian vector autoregressions. Finding good predictors for inflation: a Bayesian model averaging approach
• Abstracts: Model-based clustering of multiple time series. Binomial autoregressive moving average models with an application to US recessions
• Abstracts: Maximum-likelihood estimation of fractional cointegration with an application to U.S. and Canadian bond rates
• Abstracts: Vector autoregression modelling and forecasting. Business cycle analysis and forecasting with a structural vector autoregression model for Wales

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.