A classifying procedure for signalling turning points

Abstract

Abstract. A Hidden Markov Model (HMM) is used to classify an out of sample observation vector into either of two regimes. This leads to a procedure for making probability forecasts for changes of regimes in a time series, i.e. for turning points.Instead of maximizing a likelihood, the model is estimated with respect to known past regimes. This makes it possible to perform feature extraction and estimation for different forecasting horizons. The inference aspect is emphasized by including a penalty for a wrong decision in the cost function. The method is tested by forecasting turning points in the Swedish and US economies, using leading data. Clear and early turning point signals are obtained, contrasting favourable with earlier HMM studies.Some theoretical arguments for this are given.

Statistical surveillance of cyclical processes with application to turns in business cycles

“…For reviews and general discussions see e.g. Neftci (1982), Zarnowitz and Moore (1982), Westlund and Zackrisson (1986), Hackl and Westlund (1989), Diebold and Rudebusch (1989), Hamilton (1989), Zellner et al (1991), Jun and Joo (1993), Lahiri and Wang (1994), Li and Dorfman (1996), Koskinen and Oller (1998) and Birchenhall et al (1999).…”

“…It is similar in several aspects to e.g. the method presented by Koskinen and Oller (1998). HMM is suggested for business cycle modeling and prediction by Hamilton (1989) and is used by e.g.…”

The Application of Hidden Markov Models to the Analysis of Real Convergence

“…This approach, however, should be treated rather as a whole class of models, due to the possibility of choosing a form of an observable and unobservable component. For this reason, a huge variety of types of models were being under study over the years -see Hamilton (1994) or Koskinen and Oeller (2004) for a comprehensive review.…”

The Power of a Leading Indicators Fluctuation Trend for Forecasting Taiwans Real Estate Business Cycle: An Application of a Hidden Markov Model

“…The first part can be described as a Markov chain to generate hidden state random series; the second part of the random process is described by the distribution of the observation variable probability in the state. The basic elements are as shown below (Huang et al, 2001;Koskinen and Öller, 2004 …”