A dynamic model selection procedure to forecast using multi-process models

Article Abstract:

The use of multi-process models for forecasting is augmented by creating a dynamic procedure to evaluate the weights of the different sources, otherwise the prior probabilities of the rival models. The new standard enables the forecasting system to know more about the most logical idea for the time series, taking into account all the pooling of consecutive models to be a function of the extent of the one-step-ahead forecast error. The dynamic selection approach is used on the CP6 dataset which demonstrated a marked gain in the general predictive capacity of multi-process models whenever irregular observations arise.

Bayesian statistical decision theory, Bayesian analysis, Robust statistics, Robustness (Statistics)

---

An improved approach for estimating the mean and standard deviation of a subjective probability distribution

Article Abstract:

A formula that uses the Pearson-distribution system works effectively in estimating the mean and standard deviation of the subjective probability distribution. Compared to most major formulas that use a beta distribution assumption, such as the PERT formulas and the discounted cash flow analysis, a formula that employs the Pearson-distribution assumption yields a more accurate mean and standard deviation estimates. Also, usage of five or seven Selvidge fractiles, instead of the usual three fractiles, has resulted in lower fractile-estimation error.

Analysis, Probabilities, Probability theory, Average, Mean (Statistics), Standard deviations, Standard deviation, Distribution (Probability theory)

---

Robust modelling of ARCH models

Article Abstract:

An autoregressive conditional heteroscedastic model is proposed for modelling change variances in financial time series as an alternative to least squares

Econometrics, Autoregression (Statistics), Heteroscedasticity

Similar abstracts:

• Abstracts: Consistent moment selection procedures for generalized method of moments estimation. Tests for parameter instability and structural change with unknown change point
• Abstracts: Econometric modelling for short-term inflation forecasting in the euro area. Forecasting Euro area inflation using dynamic factor measures of underlying inflation
• Abstracts: The natural projection approach to the infinite-horizon model. Transaction costs and a redundant security: Divergence of individual and social relevance
• Abstracts: Interactions of real GNP business cycles in a three-country time-series model. The use of canonical correlation analysis to identify the order of multivariate ARMA models: Simulation and application
• Abstracts: Adoption of uncertain multi-stage technology projects: a real options approach. On the Shapley-Scarf economy: the case of multiple types 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.