Potential problems in estimating bilinear time-series models

Article Abstract:

The demand to expand linear time-series methods to incorporate non-Gaussianity, non-linearity and heterogeneity has resulted in the growing use of the bilinear model, which van capture with a good degree of accuracy any well-behaved nonlinear behavior. However, it was shown that the bilinear model has some undesirable statistical characetristics. These include the bimodality of the expected likelihood function in certain regions of the parameter space. In finite samples, moreover, the distributional properties of the parameter estimates and the standard t-statistic exhibit undesirable statistical properties.

Research, Models, Nonlinear theories, Time-series analysis, Time series analysis, Parameter estimation

---

Wheat from chaff: meta-analysis as quantitative literature review

Article Abstract:

Research is presented describing the study of economic literature and the use of statistical methods as an analytical tool in assessing the results of investigative documentation.

Economic research, Study and teaching, Literature

---

Similar abstracts:

• Abstracts: Principal-agent problems in early trading companies: a tale of two firms. Price Expectations of Business Firms: Bias in the Short and Long Run
• Abstracts: Competition policy and international trade in air transportation and telecommunications services. A tale of two reforms
• Abstracts: International macroeconomic interdependence, currency substitution, and price stickiness. Price Adjustments with Stochastic Inflation
• Abstracts: International macroeconomic interdependence, currency substitution, and price stickiness. part 2 Optimal management of international reserves
• Abstracts: Technology policy in the 1990s: old trends and new realities. Trends and motives of organizational change in Hungarian industry - a synchronic view

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