Bootstrap prediction intervals for autoregression using asymptotically mean-unbiased estimators

Article Abstract:

Extensive Monte Carlo experiments are conducted by using a number of stationery and near unit-root autoaggressive (AR) models. It was found that bias-correction based on asymptotically mean-unbiased estimation substantially improves small sample properties of bootstrap prediction intervals.

Analysis, Mathematical models

---

Forecasting using the trend model with autoregressive errors

Article Abstract:

A method for forecasting time series generated by the linear trend model with autoregressive errors, allowing a possible unit root (UR), is presented.

United States, Forecasts, trends, outlooks, Forecasts and trends, Market trend/market analysis

---

Forecasting autoregressive time series with bas-corrected parameter estimators

Article Abstract:

While all the bias-removing techniques examined offer improvement to forecasts, bootstrapping is found to deliver the best results.

Australia, Estimation theory

Similar abstracts:

• Abstracts: Competition in European aviation: pricing policy and market structure. An empirical evaluation of the determinants of research joint venture formation
• Abstracts: Benefits and costs of international financial integration: theory and facts. Capital inflows and the real exchange rate: A comparative study of Asia and Latin America
• Abstracts: MCMC methods for comparing stochastic volatility and GARCH models. Predicting returns in U.S. financial sector indices
• Abstracts: Forecast evaluation with shared data sets. Statistically significant forecasting improvements: how much out-of-sample data is likely necessary?
• Abstracts: Regression models for forecasting goals and match results in association football. Univariate versus multivariate time series forecasting: an application to international tourism demand

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