Markov Switching Model Research Articles

Out-of-Sample Predictability of the Equity Risk Premium

A large set of macroeconomic variables have been suggested as equity risk premium predictors in the literature. Acknowledging the different predictability of the equity premium in expansions and recessions, this paper proposes an approach that combines equity premium forecasts from two-state regression models using an agreement technical indicator as the observable state variable. A comprehensive out-of-sample forecast evaluation exercise based on statistical and economic loss functions demonstrates the superiority of the proposed approach versus combined forecasts from linear models or Markov switching models and forecasts from machine learning methods such as random forests and gradient boosting. The parsimonious state-dependent aspect of risk premium forecasts delivers large improvements in forecast accuracy. The results are robust to sub-period analyses and different investors’ risk aversion levels.

Large Drawdowns and Long-Term Asset Management

Long-term investors are often hesitant to invest in assets or strategies prone to significant drawdowns, primarily due to the challenge of predicting these drawdowns. This study presents a multivariate Markov-switching model for small- and large-cap returns in the U.S. equity markets, demonstrating that three distinct regimes are necessary to capture the negative trends in expected returns during financial crises. Our findings indicate that this framework enhances the prediction of conditional drawdowns compared to standard alternative models of financial returns. Furthermore, out-of-sample analysis shows that investment strategies based on these predictions outperform those relying on models with one or two regimes.

Modeling structure of inflation in Türkiye: DCC-GARCH and Markov switching model

Purpose Recent surges in inflation have posed significant challenges for Türkiye, with the annualinflation rate culminating at 83.45% by the close of 2022. The purpose of the study is to take a closer look at the details behind the rising inflation trend in Türkiye. Design/methodology/approach Due to the time-varying nature of the relationship of the variables, dynamic conditional correlation-generalized autoregressive conditionally heteroscedastic (DCC-GARCH) models and the Markov switching model are used as analytical tools. Leveraging the DCC methodology proposed by Tse and Tsui (2002), this study examined time-varying correlations, while the effect of the weighted sum of past correlations was captured using the DCC-GARCH approach introduced by Engle (2002). Findings The findings from the DCC models highlight that the exchange rate plays the most pivotal role in influencing inflation, closely followed by the money supply. In addition, the Markov switching analysis, rooted in the Phillips curve concept, identified two statistically significant regimes. The results emphasize that components of the money supply and the exchange rate stand out as primary drivers of Türkiye’s heightened inflation rates. To promote sustainable development in Turkey, the Central Bank should focus on inflation targeting, managing the money supply to align with GDP growth and adopting adaptive inflation responses. Originality/value To the best of the authors’ knowledge, this paper is the first attempt to use a combination of the DCC and Markov switching models to examine Turkish inflation from December 2005 to October 2022, according to a thorough review of previous research. Such an innovative method provides a new perspective on inflationary patterns throughout this time. In addition, this study departs from traditional approaches by including money supply measures in the analysis.

Optimal asset allocation and nonlinear return predictability from the dividend-price ratio

AbstractWe study non-linear predictability of stock returns arising from the dividend-price ratio and its implications for asset allocation decisions. Using data from five countries — U.S., U.K., France, Germany and Japan — we find empirical evidence supporting non-linear and time-varying models for the equity risk premium. Building on this, we examine several model specifications that can account for non-linear return predictability, including Markov switching models, regression trees, random forests and neural networks. Although in-sample return regressions and portfolio allocation results support the use of non-linear predictability models, the out-of-sample evidence is notably weaker, highlighting the difficulty in exploiting non-linear predictability in real time.

Mitigating the choice of the duration in DDMS models through a parametric link

One of the most important hyper-parameters in duration-dependent Markov-switching (DDMS) models is the duration of the hidden states. Because there is currently no procedure for estimating this duration or testing whether a given duration is appropriate for a given data set, an ad hoc duration choice must be heuristically justified. In this paper, we propose and examine a methodology that mitigates the choice of duration in DDMS models when forecasting is the goal. The novelty of this paper is the use of the asymmetric Aranda-Ordaz parametric link function to model transition probabilities in DDMS models, instead of the commonly applied logit link. The idea behind this approach is that any incorrect duration choice is compensated for by the parameter in the link, increasing model flexibility. Two Monte Carlo simulations, based on classical applications of DDMS models, are employed to evaluate the methodology. In addition, an empirical investigation is carried out to forecast the volatility of the S&P500, which showcases the capabilities of the proposed model.

Carbon Taxation and Electricity Price Dynamics: Empirical Evidence from the Australian Market

In this paper, we study the change of Australian electricity price dynamics that was observed before, during and after the two-year period in which a Carbon Pricing Mechanism was in force. We fit a two-states Markov Switching Model, representing a high- and a low-volatility state of the world. To avoid the interference due to periodic patterns, a deseasonalization process accounting for short- and long-term seasonality is carried out prior to the study of volatility. Estimation results highlight that, during the period when the carbon tax applies, the volatility level is lower for both the states of the world. Furthermore, the persistence in the low-volatility state is increased in the presence of the carbon tax. This conclusion is particularly relevant for macroeconomic and investment considerations because the increased uncertainty in electricity prices can significantly influence firms’ investment decisions and shape inflation expectations.

Adaptive Risk Parity Strategies for Managing Portfolio Risk in Volatile Markets

This paper presents a novel model for Adaptive Risk Parity Portfolio Construction, designed to dynamically adjust portfolio weights based on changing market conditions and economic regimes. Traditional portfolio strategies, such as Modern Portfolio Theory (MPT), often rely on static assumptions that can lead to suboptimal performance in volatile markets. In contrast, the adaptive risk parity approach aims to equalize risk contributions across asset classes while responding to fluctuations in economic indicators, such as GDP growth and inflation rates. By employing techniques such as Markov switching models and machine learning for regime identification, the proposed model allows for real-time adjustments to asset allocations. The back-testing results demonstrate that the adaptive risk parity portfolio outperforms static risk parity and traditional mean-variance optimized portfolios in terms of risk-adjusted returns and drawdown management. This research contributes to the field of portfolio management by providing a framework that enhances resilience and adaptability in investment strategies.

Modelling and forecasting European Airbnb occupancy during the pandemic: The specific merits of panel-data and Markov-switching models

This study models European Airbnb occupancy from January 2017 to December 2020 for 43 European countries within an econometric panel-data framework. Besides typical economic covariates, the COVID-19 Stringency Index is employed to allow for the recent pandemic. The counterfactual observations created by two versions of the econometric model (one with and one without European aggregates of macroeconomic controls) are benchmarked against traditional univariate time-series models and a seasonal Markov-switching autoregression. For pandemic times, the latter performs very well in terms of several forecast accuracy measures due to its ability of (a) detecting the pandemic-induced structural break and (b) accounting for seasonal patterns in the data. The econometric models also predict well as they outperform the time-series models in the majority of cases. Hence, the panel-data approach is suitable when looking for economically interpretable Airbnb occupancy forecasts, even during the time of the pandemic.

Forecasts of the mortality risk of COVID-19 using the Markov-switching autoregressive model: a case study of Nigeria (2020–2022)

The global pandemic due to SARS-Cov-2 ravaged the world and killed more than 6 million people globally within two years. Studies predicting future occurrences are essential to effectively combat the virus. This study modeled daily fatality rate in Nigeria from March 23, 2020 to March 19, 2022 and forecast future occurrences using Markov switching model (MSM). MSM estimates segmented fatality rates into three states of low-, medium- and high-risks. Further, estimates revealed that as at 19th March, 2022, Nigeria remained at the low-risk regime in which 1 (95%CI: 0, 1) person, on the average, died of coronavirus daily; however, the most probable scenario in the nearest future was the medium-risk state in which an average of 4 (95%CI: 2, 5) persons would die daily with 48.7% probability. The study concluded that Nigerian COVID mortality risks followed a switching pattern which fluctuated within low-, medium- and high-risks; however, the medium-risk state was most likely in the future. Our results indicated that the quarantine measures adopted by the governments yielded positive results. It also underscored the need for governments and individuals to intensify efforts to ensure that the country remained at the low-risk zone till the virus would be eventually eradicated.

Debt Dynamics and Fiscal Sustainability in Nigeria

This study investigated the stabilizing effect of sustainable fiscal policy on public debt dynamics as well as the impact of debt dynamics on fiscal sustainability in Nigeria for the period (1980 – 2022). Using secondary data obtained from the Central Bank of Nigeria Statistical Bulletin (2023), World Development Indicators (WDI, 2023), Debt Management Office (DMO – various years), Penn World Table (PWT – multiple years), and World Economic Outlook (WEO – various years), the study used suitable cointegration and other econometrics techniques viz: Dynamic Ordinary Least Squares (DOLS), Markov-switching model and polynomial models. It was found, amongst other things, that (i) Automatic debt dynamics can influence debt dynamics, (ii) The robustness or otherwise of primary fiscal balance determines the direction of changes in the total debt stock, and (iii) positive debt dynamics vis-à-vis dwindling revenue flows and growing government expenditure, would result in fiscal unsustainability. It was recommended, amongst other things, that since automatic debt dynamics is one of the significant influences on debt dynamics and is being fuelled by the direction of adjustments in inflation and interest rate, there is a need for concerted efforts at stabilising inflation and interest rate to stabilise public debt dynamics. Also, the study revealed that the size of the primary balance influences the changes in public debt stock. It was recommended that the authorities should endeavour to maintain a robust primary fiscal surplus to stabilise growth in public debt, if not minimise it and grow the economy.

Seasonality patterns in LNG shipping spot and time charter freight rates

The aim of this paper is to investigate the existence and the nature of seasonality in LNG freight rates of different duration contract, over different market conditions (peak and troughs) for the period from December 2010 to June 2023. We employ the HEGY method and seasonal dummy variables to test for stochastic and deterministic seasonality, respectively. Then we use Markov Switching models to test for asymmetries in seasonal fluctuations across different market conditions. We reject the existence of stochastic seasonality for all freight series while results on deterministic seasonality indicate increases in rates in June, October, and November. We also found that seasonal patterns vary across market conditions, revealing that seasonal rate movements are more pronounced when the market is in downturn. Moreover, we found that the seasonal movements present similar patterns across different trading routes. The results have implications for stakeholders across the LNG value chain.

The European Renewable Energy Sector in Calm and Turmoil Periods: The Key Role of Sovereign Risk

This paper explores the comparative role of sovereign default risk and several high-frequency macrofinancial indicators that may explain the drop in European renewable energy stocks observed during the 2008 financial crisis and the European debt crisis. We use a two-state time-varying transition probability Markov-switching model to investigate how they impact the bull and bear market trends of renewable stocks. Our main finding is that public financing conditions, captured by sovereign default risks, play a key role in both market regimes, while the other variables affect the renewable energy stocks only in calm or turmoil periods. Moreover, sovereign risk is identified as the main determinant of the European renewable energy stock dynamics in both regimes in the period under review. Finally, we suggest that this effect may be due to the sensitivity of investors to the energy policy uncertainty, entailed by such a pressure on public finances. JEL Classification: G01, G15, Q42

Well‐posedness and error analysis of wave equations with Markovian switching

Compared to traditional partial differential equation modeling methods, Markov switching models can accurately capture the abrupt changes or jumps that complex systems often experience in the real world. In this paper, we propose a novel wave equation model with Markovian switching to represent complex systems with state‐jumping phenomena better, and the well‐posedness of the model is proved. In addition, a numerical method with non‐uniform grids is also proposed for the proposed model to simulate the data in realistic situations, which is based on the use of finite element discretization in space and central difference discretization in time. Finally, we conduct several experiments to analyze the errors and stability of the proposed model and the traditional model. The results show that the Markov switching model proposed in this paper has a smaller error than the traditional models while ensuring stability and can more accurately simulate the state jump phenomena of real‐world systems.

Zika emergence, persistence, and transmission rate in Colombia: a nationwide application of a space-time Markov switching model.

Zika, a viral disease transmitted to humans by Aedes mosquitoes, emerged in the Americas in 2015, causing large-scale epidemics. Colombia alone reported over 72,000 Zika cases between 2015 and 2016. Using national surveillance data from 1121 municipalities over 70 weeks, we identified sociodemographic and environmental factors associated with Zika's emergence, re-emergence, persistence, and transmission intensity in Colombia. We fitted a zero-state Markov-switching model under the Bayesian framework, assuming Zika switched between periods of presence and absence according to spatially and temporally varying probabilities of emergence/re-emergence (from absence to presence) and persistence (from presence to presence). These probabilities were assumed to follow a series of mixed multiple logistic regressions. When Zika was present, assuming that the cases follow a negative binomial distribution, we estimated the transmission intensity rate. Our results indicate that Zika emerged/re-emerged sooner and that transmission was intensified in municipalities that were more densely populated, at lower altitudes and/or with less vegetation cover. Warmer temperatures and less weekly-accumulated rain were also associated with Zika emergence. Zika cases persisted for longer in more densely populated areas with more cases reported in the previous week. Overall, population density, elevation, and temperature were identified as the main contributors to the first Zika epidemic in Colombia. We also estimated the probability of Zika presence by municipality and week, and the results suggest that the disease circulated undetected by the surveillance system on many occasions. Our results offer insights into priority areas for public health interventions against emerging and re-emerging Aedes-borne diseases.

Do Exchange Rates Respond Asymmetrically to Crude Oil Market Shocks? Insights from BRICS and Pakistan

Abstract The shocks in the crude oil market are unlikely to have symmetric impacts on exchange rates, especially over various time periods. Furthermore, in low- and high-volatility regimes, exchange rates are predicted to respond differently to oil market shocks. By employing Markov-switching model and state-space model, this research empirically investigates whether oil market shocks exert asymmetric impacts on exchange rates, across the time and across different levels of volatility. In doing so, the study utilizes monthly data from January 1994-September 2017 for BRICS countries and Pakistan. The findings indicate that although oil supply shocks are mostly insignificant, oil price shocks appear statistically significant to affect the exchange rates whereas aggregate demand shocks are most volatile and are more likely to cause exchange rate fluctuations. The study concludes that various forms of oil shocks have a wide range of consequences on exchange rate determinations. Nonetheless, exchange rates’ response to oil shocks differ dramatically through low- and high-volatility states, implying that oil shocks exert a time-varying impact on exchange rates. Finally, the findings provide credence to the phenomenon of asymmetry in exchange rate responses to oil market shocks.

Monetary Policy Spillovers and Inter-Market Dynamics Perspective of Preferred Habitat Model

This study advances the understanding of the Preferred Habitat Model’s capacity to shed light on the inter-market transfer of mean returns and the diffusion of price volatility in Pakistani investment markets. It examines the extent to which returns in one market exert a systematic influence on returns across others under the potential sway of interest rate policy shifts, USD exchange rate volatility, and domestic inflation trends. Employing a methodological arsenal that includes the GARCH process, enhanced by Dynamic Conditional Correlations (DCC), as well as the Markov Switching Model, this research assesses the propagation of mean returns and volatility across markets. The analysis uncovers significant linkages between monetary policy and stock market indices, underscoring the profound impact of monetary policy on cross-market performance transmission. These insights are pivotal for regulators overseeing the nuanced interaction between monetary policy and market performance. They are crucial for local and international investors interested in developing economies, especially in Pakistan’s markets.

Optimal risk management considering environmental and climatic changes.

Climate change presents challenges to policy and economic stability, necessitating effective trading strategies to reduce environmental risks. This article addresses gaps in existing studies by using a Markov-switching model to consider climate risk. Backward stochastic differential equationsare used to optimize utility with three hedging strategies based on the concept of risk aversion. Numerical scenarios confirm the model's superiority in incorporating exogenous events, with our risk-averse strategy outperforming classical approaches. Our strategy outperforms classical strategies by taking a flexible risk trading when investors face risk-averse behavior due to climate risk events. The findings presented in this article have important implications for the development of more resilient investment portfolios and can contribute to climatepolicy.

Bayesian Markov switching model for BRICS currencies' exchange rates

AbstractExchange rate modeling has always fascinated researchers because of its complex macroeconomic dynamics. This study documents the exchange rate dynamics of major emerging economies after accounting for their macroeconomic cycles and explores the Bayesian Vector Error Correction Model (VECM) Markov Regime switching model, which uses time‐varying transition probabilities. The main objective is to study the exchange rate dynamics of Brazil, Russia, India, China, and South Africa (BRICS) vis‐à‐vis the US dollar. The Bayesian setup uses two hierarchal shrinkage priors, the normal‐gamma (NG) prior and the Litterman prior, for parameters' estimation. These shrinkage priors allow for a more comprehensive assessment of the regime‐specific coefficients. The model performed well in differentiating between the two regimes for all currencies. The Russian ruble was identified to be the most depreciated currency, whereas the African Rand was the most appreciated. The evaluation of model features revealed that many regime‐specific coefficients differed significantly from their common mean. A forecasting exercise was then performed for the out‐of‐sample period to assess the model's performance. A significant improvement was observed over the basic random walk (RW) model and the linear Bayesian vector autoregression (BVAR) model.

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.

Fiscal regimes and debt sustainability in Colombia

ABSTRACT This paper evaluates Colombia’s debt sustainability for the period 1980–2021, using a nonlinear fiscal reaction function approach. We employ a Markov-Switching model with annual data to identify sustainable and unsustainable fiscal regimes. According to this model, we identified an unsustainable fiscal regime during three periods with an average duration of six years − 1980–1985, 1993–1999, and 2015–2021 – while the other periods are identified as a sustainable regime. Although we find evidence of asymmetric behaviour in fiscal responses and the presence of pro-cyclical fiscal policy, Colombia’s long-term fiscal sustainability is verified based on a globally-based fiscal sustainability test. In this context, the government reacts appropriately to increases in debt only in the sustainable regime, but the fiscal response is sufficiently robust to stabilise public debt throughout the entire period. Finally, we also highlight concerns regarding the current state of the global economy, particularly due to an inflationary trend and escalating interest rates, which can compromise long-term fiscal sustainability.