Combination of forecasts using self-organizing algorithms

Article Abstract:

By combining different forecasting models in an appropriate way we can achieve forecast accuracy and improve overall performance of forecasts. Linear ones and nonlinear ones are the two methods for combining forecasting methods. Self-organizing combination forecasting method has improved the forecasting capability by using the information directly, unlike neural network method.

China, Data warehousing/data mining, Data mining

---

The multi-chain Markov switching model

Article Abstract:

A new multi-chain Markov switching model for forecasting is presented and examined. The transition probabilities of this model are qualified by the reliance on the regime of the other variables. Regime switching, multivariate time series, etc. are also discussed.

Europe, Markov processes

---

A note on in-sample and out-of-sample tests for Granger causality

Article Abstract:

In-sample and out-of-sample tests for Granger causality are studied with the help of Monte Carlo simulation. Out-of-sample tests are established as more potent and useful than in-sample tests.

United Kingdom

Similar abstracts:

• Abstracts: A forecasting comparison of classical and Bayesian methods for modelling logistic diffusion. A cautionary note on outlier robust estimation of threshold modelsn
• Abstracts: The maximum and minimum of primary forecasts. The uses and abuses of 'consensus' forecasts
• Abstracts: A fractal forecasting model for financial time series. The importance of interest rates for forecasting the exchange rate
• Abstracts: The internal distribution of union rents: an empirical test of the voting power model. The impact of real rents and wages on household formation
• Abstracts: Finite sample behavior of tests for grouped heteroskedasticity. Finding cointegration rank in high dimensional systems using the Johansen test: an illustration using data based Monte Carlo simulations

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