Univariate Markov-switching Models Research Articles

Identification and forecasting of bull and bear markets using multivariate returns

SummaryBull and bear market identification generally focuses on a broad index of returns through a univariate analysis. This paper proposes a new approach to identify and forecast bull and bear markets through multivariate returns. The model assumes that all assets are directed by a common discrete state variable from a hierarchical Markov switching model. The hierarchical specification allows the cross‐section of state‐specific means and variances to differ over bull and bear markets. We investigate several empirically realistic specifications that permit feasible estimation even with 100 assets. Our results show that the multivariate framework provides competitive bull and bear regime identification and improves portfolio performance and density prediction compared with several benchmark models including univariate Markov switching models.

Nonlinear relationship between monetary policy and stock returns: Evidence from the U.S.

This paper studies the relationship between monetary policy and stock market return in the U.S. using nonlinear econometric models. It first employs a univariate Markov-switching model on each of the three stock indices and three monetary policy variables, displaying significant regime-switching patterns and common movements. This paper then uses a Markov-switching dynamic bi-factor model to simultaneously extract two latent common factors from stock indices and monetary policy variables to represent monetary policy changes and stock market movements separately. The smoothed probabilities of regimes demonstrate that expansionary monetary policy regimes follow economic recessions, but bear stock markets usually occur before economic recessions. The maximum likelihood estimation results show that expansionary monetary policy such as a decrease in the federal funds rate raises stock returns, but stock returns don't directly influence monetary policy decision.

Identification and Real-Time Forecasting of Norwegian Business Cycles

We define and forecast classical business cycle turning points for the Norwegian economy. When defining reference business cycles, we compare a univariate and a multivariate Bry-Boschan approach with univariate Markov-switching models and Markov-switching factor models. On the basis of a receiver operating characteristic curve methodology and a comparison of business cycle turning points with Norway’s main trading partners, we find that a Markov-switching factor model provides the most reasonable definition of Norwegian business cycles for the sample 1978Q1-2011Q4. In a real-time out-of-sample forecasting exercise, focusing on the last recession, we show that univariate Markov-switching models applied to surveys and a financial conditions index are timely and accurate in calling the last peak in real time. The models are less accurate and timely in calling the trough in real time.

Identification and real-time forecasting of Norwegian business cycles

We define and forecast classical business cycle turning points for the Norwegian economy. When defining reference business cycles, we compare a univariate and a multivariate Bry–Boschan approach with univariate Markov-switching models and Markov-switching factor models. On the basis of a receiver operating characteristic curve methodology and a comparison of the business cycle turning points of Norway’s main trading partners, we find that a Markov-switching factor model provides the most reasonable definition of Norwegian business cycles for the sample 1978Q1–2011Q4. In a real-time out-of-sample forecasting exercise, focusing on the last recession, we show that univariate Markov-switching models applied to surveys and a financial conditions index are timely and accurate in calling the last peak in real time. However, the models are less accurate and timely in calling the trough in real time.

The Effects of Oil Price Shocks on Transitional Dynamics of Turkish Business Cycle

In this study we investigated usefulness of oil price shocks in predicting switches between the growth phases of output in Turkey using monthly data for the period 1986-2014 by extending Markov Switching framework to include time-varying transition probabilities. We investigate the issue of whether the addition of various real oil price shocks to a univariate Markov Switching model for output can characterize the dynamics of business cycles better than the fixed transition probability version of Markov Switching model. The main results are summarized as follows. We find that although information about the lags of output growth and the information contained in transition probabilities combine to help identify which state of the economy has occurred in the TVTP model, oil price shocks are not the leading indicator of Turkish business cycle.

Spillover effect of US monetary policy to ASEAN stock markets: Evidence from Indonesia, Singapore, and Thailand

In this paper, we investigate the spillover effect from US monetary policy to selected ASEAN stock markets by employing Markov-switching models. Based on univariate Markov-switching models, we confirm the existence of two distinct regimes for both US monetary policy and the stock markets. By applying multivariate Markov-switching models, we find that US interest rates have a negative effect on the selected ASEAN stock markets during economic expansion periods. However, this kind of effect disappears during economic crisis periods. Our empirical results indicate that the spillover effect from US monetary policy influences the ASEAN stock markets only during the tranquil period. These results have important implications for the transmission mechanisms of asset price, such as the credit channel, trade channel, and balance sheet channel.

Measuring and predicting heterogeneous recessions

This paper examines the usefulness of a more refined business cycle classification for monthly industrial production (IP), beyond the usual distinction between expansions and contractions. Univariate Markov-switching models show that a three regime model is more appropriate than a model with only two regimes. Interestingly, the third regime captures ‘severe recessions’, contrasting the conventional view that the additional third regime represents a ‘recovery’ phase. This is confirmed by means of Markov-switching vector autoregressive models that allow for phase shifts between the cyclical regimes of IP and the Conference Board's Leading Economic Index (LEI). The timing of the severe recession regime mostly corresponds with periods of substantial financial market distress and severe credit squeezes, providing empirical evidence for the ‘financial accelerator’ theory.

Measuring and Predicting Heterogeneous Recessions

This paper examines whether the Conference Board's Leading Economic Index (LEI) can be used for modeling and forecasting a more refined business cycle classification beyond the usual distinction between expansions and contractions. Univariate Markov-switching models for monthly coincident variables and the LEI show that a three regime model is more appropriate than a model with only two regimes. Interestingly, the third regime captures 'severe recessions' contrasting the conventional view that the additional third regime represents a 'recovery' phase. This is confirmed by means of Markov-switching vector autoregressive models that allow for phase shifts between the cyclical regimes of LEI and industrial production. Results indicate that a three regime model with a severe recession phase describes the cyclical dynamics in these series better than a two regime model (with only recession and expansion regimes) and a three regime model with a recovery phase. The timing of the third regime mostly corresponds with periods of substantial credit squeezes and dramatic increases in the default spread as in the recent recession of 2007-2009. These findings provide empirical evidence for the theory of 'financial accelerator'. The severe recession regime of the LEI leads that of IP by 6.5 months whereas for mild recessions this lead time increases to one year.

Incorporating a leading indicator into the trading rule through the Markov-switching vector autoregression model

This article examines the profitability of trading rules based on the smoothed probability of Markov-switching models and executes two models in Taiwan's case. The results present that both proposed models can earn excess returns over the buy-and-hold strategy and support that both can be used to trade. However, the univariate Markov-switching model, which only uses daily returns series does not successfully capture the trend in the stock market, especially during a bull market. This implies that high-frequency returns series contain lots of noises. In order to overcome this problem, the Markov-switching vector autoregression model that combines a leading indicator and returns is performed in this study. The results indicate a better trading pattern. We conclude that the leading indicator chosen from open interest in the future market increases useful information and reduces noises to improve model estimation, which can well identify the position of bull and bear markets.

Markov-switching and the Beveridge–Nelson decomposition: Has US output persistence changed since 1984?

Abstract We show that, for a class of univariate and multivariate Markov-switching models, exact calculation of the Beveridge–Nelson (BN) trend/cycle components is possible. The key to exact BN trend/cycle decomposition is to recognize that the latent first-order Markov-switching process in the model has an AR(1) representation, and that the model can be cast into a state-space form. Given the state-space representation, we show that impulse-response function analysis can be processed with respect to either an asymmetric discrete shock or to a symmetric continuous shock. The method presented is applied to Kim, Morley, Piger’s [Kim, C.-J., Morley, J., Piger, J., 2005. Nonlinearity and the permanent effects of recessions. Journal of Applied Econometrics 20, 291–309] univariate Markov-switching model of real GDP with a post-recession ‘bounce-back’ effect and Cochrane’s [Cochrane, J.H., 1994. Permanent and transitory components of GNP and stock prices. Quarterly Journal of Economics 109, 241–263] vector error correction model of real GDP and real consumption extended to incorporate Markov-switching. The parameter estimates, the BN trend/cycle components, and the impulse-response function analysis for each of these empirical models suggest that the persistence of US real GDP has increased since the mid-1980’s.

The asymmetric effects of oil shocks on output growth: A Markov–Switching analysis for the G-7 countries

Oil shocks are generally acknowledged to have important effects on both economic activity and macroeconomic policy. The aim of this paper is to investigate how oil price shocks affect the growth rate of output of a subset of developed countries by comparing alternative regime switching models. Different Markov–Switching (MS) regime autoregressive models are, therefore, specified and estimated. In a successive step, univariate MS models are extended in order to verify if the inclusion of asymmetric oil shocks as an exogenous variable improves the ability of each specification to identify the different phases of the business cycle for each country under scrutiny. Following the wide literature on this topic, seven different definitions of oil shocks which are able to describe oil price changes, asymmetric transformations of oil price changes, oil price volatility, and oil supply conditions are considered. Our findings can be summarized as follows. While the introduction of different oil shock specifications is never rejected, positive oil price changes, net oil price increases and oil price volatility are the oil shock definitions which contribute to a better description of the impact of oil on output growth. In addition, models with exogenous oil variables generally outperform the corresponding univariate specifications which exclude oil from the analysis. However, a stability analysis of the coefficients across different subsamples shows that the role of oil shocks in explaining recessionary episodes has changed over time. Improvements in energy efficiency, together with a better systematic approach to external supply and demand shocks by monetary and fiscal authorities are argued to be responsible for the changing macroeconomic effects of oil shocks. Finally, the impact of G-7 countries aggregate growth on oil market conditions is considered and assessed empirically. The null hypothesis of the absence of a reverse relationship from real GDP growth to oil prices is rejected by the data.

An analysis of regime shifts in the Turkish economy

We use a time-varying dynamic factor model with regime switching to construct and estimate the leading indicators of the currency crises in Turkey. After that, we analyze the business cycles of the Turkish economy, by using a three-state univariate Markov-switching model. Both models capture the observed dynamics of the Turkish economy over the period 1987–2002.

Inflation Transmission in the EMU: A Markov-Switching VECM Analysis

This paper analyzes the transmission of inflation across the five largest economies in the European Monetary Union, i.e. France, Germany, Italy, Netherlands and Spain. We use monthly CPI inflation rates for the period 1970-2006. Given the long observation period and the continuing economic integration of Europe’s economies, we first try to investigate, if there were changes in inflation dynamics in these countries using univariate Markov-switching models. To assess the inflation transmission mechanism, we first establish a long-run relationship between the five countries using cointegration methods. As implied by the results of the univariate models, we allow for changes in the adjustment coefficients of the cointegrating relationships and the short-run dynamics. Using a Markov-switching vector error correction model we find evidence for multiple regime switches during the early 1970s till the mid 1980s. Exactly during this period we find evidence for Germany being weakly exogenous, which highlights the dominance of German monetary policy at this time. Since the mid-1980s we find evidence for a stable transmission mechanism both in the long- and the short-run characterized by a low degree of inflation persistence.

Regime switching and artificial neural network forecasting of the Cyprus Stock Exchange daily returns

AbstractThis paper provides an analysis of regime switching in volatility and out‐of‐sample forecasting of the Cyprus Stock Exchange by using daily data for the period 1996–2002. We first model volatility regime switching within a univariate Markov switching framework. Modelling stock returns within this context can be motivated by the fact that the change in regime should be considered as a random event and not predictable. The results show that linearity is rejected in favour of an MS specification, which forms statistically an adequate representation of the data. Two regimes are implied by the model, the high‐volatility regime and the low‐volatility one, and they provide quite accurately the state of volatility associated with the presence of a rational bubble in the capital market of Cyprus. Another implication is that there is evidence of regime clustering. We then provide out‐of‐sample forecasts of the CSE daily returns by using two competing nonlinear models, the univariate Markov switching model and the Artificial Neural Network Model. The comparison of the out‐of‐sample forecasts is done on the basis of forecast accuracy, using the Diebold and Mariano test and forecast encompassing, using the Clements and Hendry test. The results suggest that both nonlinear models are equivalent in forecasting accuracy and forecasting encompassing, and therefore on forecasting performance. Copyright © 2006 John Wiley & Sons, Ltd.

Regime Switching and Artificial Neural Network Forecasting of the Cyprus Stock Exchange Daily Returns

This paper provides an analysis of regime switching in volatility and out-of-sample forecasting of the Cyprus Stock Exchange using daily data for the period 1996-2002. We first model volatility regime switching within a univariate Markov-Switching framework. Modelling stock returns within this context can be motivated by the fact that the change in regime should be considered as a random event and not predictable. The results show that linearity is rejected in favour of a MS specification, which forms statistically an adequate representation of the data. Two regimes are implied by the model; the high volatility regime and the low volatility one and they provide quite accurately the state of volatility associated with the presence of a rational bubble in the capital market of Cyprus. Another implication is that there is evidence of regime clustering. We then provide out-of-sample forecasts of the CSE daily returns using two competing non-linear models, the univariate Markov Switching model and the Artificial Neural Network Model. The comparison of the out-of-sample forecasts is done on the basis of forecast accuracy, using the Diebold and Mariano (1995) test and forecast encompassing, using the Clements and Hendry (1998) test. The results suggest that both non-linear models equivalent in forecasting accuracy and forecasting encompassing and therefore on forecasting performance.

THE ARGENTINEAN CURRENCY CRISIS: A MARKOV-SWITCHING MODEL ESTIMATION

In 2002, the Argentinean currency board came to a sudden and dramatic end. Although the country had been suffering from weak economic fundamentals for years, the timing and severity of the currency crisis surprised most observers. The present study analyzes the role of fundamentals and self-fulfilling speculation in the Argentinean crisis. Arguing within a theoretical model of a fixed exchange rate system that allows for multiple equilibria, we show that the crisis, although associated with weak fundamentals, cannot be explained by these macroeconomic factors alone. Estimating a univariate Markov-switching model, the current study shows that shifts in agents’ beliefs did indeed also play a crucial role.

Can regime‐switching models reproduce the business cycle features of US aggregate consumption, investment and output?

AbstractThe ability of Markov‐switching (MS) autoregressive models to replicate selected classical business cycle features found in US post‐war consumption, investment and output is compared to that of linear models. Univariate MS models appear to offer more dynamically parsimonious representations, but generally are unable to reproduce features missed by linear models. In the multivariate models, some cointegration restrictions were found to have a crucial impact, and the ability of models that imposed cointegration to reproduce business cycle features was enhanced by Markov switching. Copyright © 2004 John Wiley & Sons, Ltd.

Multivariate Markov Switching Common Factor Models for the UK

We estimate a model that incorporates two key features of business cycles, comovement among economic variables and switching between regimes of boom and slump, to quarterly UK data for the last four decades. A common factor, interpreted as a composite indicator of coincident variables, and estimates of turning points from one regime to the other, are extracted from the data by using the Kalman filter and maximum likelihood estimation. Both comovement and regime switching are found to be important features of the UK business cycle. The composite indicator produces a sensible representation of the cycle and the estimated turning points agree fairly well with independently determined chronologies. These estimates are sharper than those produced by a univariate Markov switching model of GDP alone. A fairly typical stylized fact of business cycles is confirmed by this model – recessions are steeper and shorter than recoveries.

Is the U.S. Recession Over Yet

As of March 2002 U.S. economy is already showing signs of breaking out of 2001 However, NBER announced in February 2002 that the Committee will wait until a substantial period of expansion has elapsed before declaring that a turning point in economy is a true trough, marking end of a recession. Thus, NBER dating procedure can not be used in real time. This raises some questions: How can we assess whether a recession is over until NBER announces its decision? Are there models that can date business cycles in real time? Univariate Markov switching models and Dynamic Factor models with regime switching yield estimated probabilities or recessions and expansions that are very similar to NBER dating. Thus, these formal analytic models can be useful to monitor turning points and evaluate forecasts in real time, overcoming delays inherent to NBER dating procedure. In particular, these models can be used to assess when 2001 U.S. recession actually started and whether recession is over yet. Regarding beginning of recession, models indicate that NBER Committee was right on target -- that is, March 2001 was beginning of With respect to possible end of recession, models indicate that recession has already ended, using information available up to March 2002. A conservative estimate places trough of recession in February 2002 whereas an optimistic forecast places end of recession in January 2002.