Testing for unit roots using forward and reverse Dickey-Fuller regressions
Article Abstract:
A simple approach is introduced for testing the null hypothesis that a process has a unit root against a stationary alternative. Since the method is based on the joint values of the standard Dickey-Fuller tests when used to both the forward and reverse data realizations, computation does not need complex techniques and can be done using available econometric software packages.
Publication Name: Oxford Bulletin of Economics & Statistics
Subject: Economics
ISSN: 0305-9049
Year: 1995
User Contributions:
Comment about this article or add new information about this topic:
Specification testing of Markov switching models
Article Abstract:
A set of tests is presented to evaluate the appropriateness of Markov switching models, for both non-parametric and parametric encompassing and model consistency. Simulation is used in combination with conditional mean estimation and non-parametric density; results are discussed.
Publication Name: Oxford Bulletin of Economics & Statistics
Subject: Economics
ISSN: 0305-9049
Year: 2003
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: Accounting for income inequality in rural China: a regression-based approach. Determinants of the performance of foreign invested enterprises in China
- Abstracts: Perpetual signalling with imperfectly correlated costs. Auction form preferences of risk-averse bid takers. Revenue-raising versus other approaches to environmental protection: the critical significance of preexisting tax distortions
- Abstracts: Receding horizon control of jump linear systems and a macroeconomic policy problem. Saddle-point calculation for constrained finite Markov chains
- Abstracts: Jamming a rival's learning. Ranking auctions with risk averse bidders. Bidding for the future: signaling in auctions with an aftermarket
- Abstracts: Public pension reform in Europe: policies, prospects and evaluation. Ageing, retirement and pension reforms