A comparison of approximate Bayesian forecasting methods for non-Gaussian time series

Article Abstract:

The effectiveness of approximate Bayesian methods are compared for the analysis and forecasting of non-Gaussian dynamic processes. The Bayesian analysis of non-linear time series is carried out by developing a numerical algorithm based on Monte Carlo Markov Chain methods.

United States, Bayesian statistical decision theory, Bayesian analysis, Gaussian processes

---

ARMA models and the Box-Jenkins methodology

Article Abstract:

A study applies various aspects of Box-Jenkins methodology to Auto Regressive Moving Average (ARMA) models in order to find out why in empirical tests, post-sample forecasting accuracy are worse than simpler time series methods. Results revealed that the most important factor in determining post-sample forecasting accuracies is the way the data are made stationary in its mean. Utilizing ARMA models to seasonally adjusted data slightly enhances post-sample accuracies and simplifies the use of ARMA models.

---

Observability of states in non-linear systems

Article Abstract:

Two new properties, namely linear observability and L-observability, were used to help in the examination of overspecification and unidentifiability in quasi-linear and non-linear state space time series. Numerous equivalence outcomes are established and several general properties of observable and unobservable methods are obtained. The outcomes are mostly established by employing graphical methods.

Research and Development in the Physical, Engineering, and Life Sciences, Statistics, Mathematical models, Graphic methods, Statistics (Mathematics)

Similar abstracts:

• Abstracts: Intertemporal recursive utility and an equilibrium asset pricing model in the presence of Levy jumps. Properties of bid and ask reservation prices in the rank-dependent expected utility model
• Abstracts: Covariance matrix estimation and the power of the overindentifying restrictions test. Incomplete learning from endogenous data in dynamic allocation

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