A BIVARIATE AUTOREGRESSIVE PROBIT MODEL: BUSINESS CYCLE LINKAGES AND TRANSMISSION OF RECESSION PROBABILITIES. (18th March 2013)

• Record Type: Journal Article
• Authors: Nyberg, Henri
• Is Part Of: Macroeconomic dynamics. Volume 18:Number 4(2014)
• Page Start: 838
• Page End: 862
• Publication Date: 2013-03-18
• Subjects: Macroeconomics -- Periodicals; 339.05
• DOI: 10.1017/S1365100512000636
• Languages: English
• ISSNs: 1365-1005

Abstract

<abstract abstract-type="normal"> <title> <x content-type="archive" xml:space="preserve">Abstract</x> </title> <p>I propose a new binary bivariate autoregressive probit model of the state of the business cycle. This model nests various special cases, such as two separate univariate probit models used extensively in the previous literature. The parameters are estimated by the method of maximum likelihood and forecasts can be computed by explicit formulae. The model is applied to predict the U.S. and German business cycle recession and expansion periods. Evidence of in-sample and out-of-sample predictability of recession periods by financial variables is obtained. The proposed bivariate autoregressive probit model allowing links between the recession probabilities in the United States and Germany turns out to outperform two univariate models.</p> </abstract>