Finite-sample properties of tests for equal forecast accuracy

Article Abstract:

Issues concerning the effects of small samples on statistical forecasting are discussed. By means of Monte Carlo experiments, Diebold-Mariano, Morgan-Granger-Newbold and Christiano tests are analysed. Results show that all the tests suffer from size distortions.

Administration of General Economic Programs, Economic Statistics & Research, Usage, Economic conditions, Economic forecasting, Statistics, Statistics (Data), Statistical sampling, Sampling (Statistics)

---

Can out-of-sample forecast comparisons help prevent overfitting?

Article Abstract:

The effectiveness of out-of-sample forecast comparisons in preventing data mining-induced overfitting in finite samples is examined. Simulation results show that out-of-sample forecast comparisons can help avoid overfitting.

Comparative analysis

---

Bias-corrected bootstrap prediction regions for vector autoregression

Article Abstract:

Performance of alternative bias-corrected bootstrap prediction regions for vector autoregressive model is examined. Bootstrap prediction regions based on asymptotic bias-correction show better small sample properties.

Vector analysis

Similar abstracts:

• Abstracts: The canonical extensive form of a game form. Part II. Representation. Stable voting procedures for committees in economic environments
• Abstracts: Von Neumann Morgenstern preferences. Harsanyi's social aggregation theorem for state-contingent alternatives. An extension of a theorem of von Neumann and Morgenstern with an application to social choice theory
• Abstracts: Interim core concepts for a bayesian pure exchange economy. Equitable allocation of divisible goods. Market allocation of indivisible goods

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