Detecting Economic Regimes in France : a Qualitative Markov-Switching Indicator Using Mixed Frequency Data

This paper proposes an indicator for detecting business cycle turning points involving mixed frequency business survey unbalanced data. It is based on a hidden Markov-switching model and allows for the detection of regime changes in a given economy where information is displayed monthly and/or quarterly. Starting from Gregoir and Lenglart (2000) we propose an adaptable framework which can be applied to many situations involving monthly, bimonthly and quarterly data. The proposed methodology is applied to the French economy. Using balances from business survey, this indicator measures the probability of being in an accelerating or a decelerating phase referring to the output growth rate cycle. The index is confronted over the past with a reference dating based on the growth cycle of the French GDP estimated through a Christiano-Fitzgerald filter. By extracting information from business survey, our index exhibits quite clearly and timely regime changes in France. Moreover, the signal delivered by the indicator is mainly unrevised and available many quarters before the ex-post dating. Considering this adequacy with the reference dating over the past, the turning point index therefore provides an accurate signal on the current outlook.Business Cycle; Business Survey; Turning points; Markov Switching Indicator; Multifrequency Data