Evaluating the predicting power of ordered probit models for multiple business cycle phases in the U.S. and Japan
| Autor: | Christian R. Proaño, Artur Tarassow |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Economics and Econometrics
Computer science media_common.quotation_subject 05 social sciences Ordered probit Recession Set (abstract data type) Identification (information) Economic indicator Probit model 0502 economics and business Political Science and International Relations Business cycle Econometrics 050207 economics Finance Selection (genetic algorithm) 050205 econometrics media_common |
| Zdroj: | Journal of the Japanese and International Economies. 50:60-71 |
| ISSN: | 0889-1583 |
| DOI: | 10.1016/j.jjie.2018.08.002 |
| Popis: | We investigate the probability forecasting performance of a three-regime dynamic ordered probit econometric framework suitable to forecast recessions, low growth periods and accelerations for the U.S. and Japan. In a first step, we apply a non-parametric dating algorithm for the identification of these three phases along the lines of Proano (2017). We compare the pseudo-out-of-sample forecasting skills of an otherwise standard binary dynamic probit model with a three-regime dynamic ordered probit framework by means of a rolling-window exercise combined with time-varying indicator selection. Based on a set of monthly macroeconomic and financial leading indicators, the results show the superiority of the ordered probit framework to forecast all three business cycle phases up to six months ahead under real-time conditions. Receiver operating characteristics and related summarizing statistics are applied as probability forecast evaluation measures. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect