Time-varying Parameter Vector Autoregression Research Articles

Bayesian analysis of time-varying interactions between stock returns and foreign equity flows

This study discusses the trading behavior of foreign investors with respect to economic uncertainty in the South Korean stock market from a time-varying perspective. We employ a news-based measure of economic uncertainty along with the model of time-varying parameter vector autoregression with stochastic volatility. The empirical analysis reveals several new findings about foreign investors’ trading behaviors. First, we find evidence that positive feedback trading often appears during periods of high economic uncertainty, whereas negative feedback trading is exclusively observable during periods of low economic uncertainty. Second, the foreign investors’ feedback trading appears mostly to be well-timed and often leads the time-varying economic uncertainty except in periods of global crises. Third, lagged negative (positive) response of net flows to economic uncertainty is found to be coupled with lagged positive (negative) feedback trading. Fourth, the study documents an asymmetric response of foreign investors with regard to negative and positive shocks of economic uncertainty. Specifically, we find that they instantly turn to positive feedback trading after a negative contemporaneous response of net flows to shocks of economic uncertainty. In contrast, they move slowly toward negative feedback trading after a positive response of net flows to uncertainty shocks.

Dynamic spillovers between energy and stock markets and their implications in the context of COVID-19

This study combined time-varying parameter vector autoregression (TVP-VAR) and a spillover index model to analyze the static, total, and net spillover effects of energy and stock markets before and after the COVID-19 outbreak. A network method was also used to depict structural changes more intuitively. Furthermore, we calculated and compared changes in the hedge ratio, optimal portfolio weights, and hedge effectiveness to guide investors to adjust portfolio strategies during COVID-19. The main findings were as follows: First, COVID-19 had a significant impact on spillover effects, and the average value of total spillover index increased by 19.94% compared with that before the epidemic. Second, the energy market was an important risk recipient of the stock market before COVID-19, and the extent of risk acceptance increased after the COVID-19 outbreak. Third, the hedging ratio, optimal portfolio weights, and hedge effectiveness showed huge changes after the COVID-19 outbreak, requiring investors to adjust their portfolio strategies.

Feverish sentiment and global equity markets during the COVID-19 pandemic

This paper proposes a new approach to estimating investor sentiments and their implications for the global financial markets. Contextualising the COVID-19 pandemic, we draw on the six behavioural indicators (media coverage, fake news, panic, sentiment, media hype and infodemic) of the 17 largest economies and data from 1st January 2020 to 3rd February 2021. Our key findings, obtained using a time-varying parameter-vector auto-regression (TVP-VAR) model, indicate the total and net connectedness for the new index, entitled ‘feverish sentiment’. This index provides us insight into economies that send or receive the sentiment shocks. The construction of the network structures indicates that the United Kingdom, China, the United States and Germany became the epicentres of the sentimental shocks that were transmitted to other economies. Furthermore, we also explore the predictive power of the newly constructed index on stock returns and volatility. It turns out that investor sentiment positively (negatively) predicts the stock volatility (return) at the onset of COVID-19. This is the first study of its kind to assess international feverish sentiments by proposing a novel approach and its impacts on the equity market. Based on empirical findings, the study also offers some policy directions to mitigate the fear and panic during the pandemic.

Research on the Effectiveness of China’s Macro Control Policy on Output and Technological Progress under Economic Policy Uncertainty

The uncertainty of economic policy, a specific form of uncertainty, can affect both economic growth, and the effectiveness of the macroeconomic regulation and control policy. Existing studies have analyzed the impacts of economic policy uncertainty on investment, consumption, trade, and total factor survival, but there is no analysis of the effectiveness of macroeconomic regulation and control policies on output and technological progress in a deterministic environment. Output growth and technological progress show the performance of economic growth in gross and efficiency, respectively, which is the external performance and internal driving force of economic growth. To achieve long-term sustainable economic development, it is necessary to consider both the aggregate problem and technological progress. In this context, this paper attempts to explore the effectiveness of China’s macroeconomic regulation and control policy on output growth and technological progress under the economic policy uncertainty. Specifically, this paper first analyzes the effectiveness of macroeconomic regulation and control policy on China’s output growth and technological progress in an uncertain environment, and then makes an empirical study by constructing a time-varying parameter vector autoregression model (TVP-VAR). Furthermore, the simulation test of the relevant results is carried out using the counter-fact analysis method. The empirical results show that: (1) under the uncertainty environment, the direction of the effect of price monetary policy on output has not changed, the effect of interest rate increase on output growth is negative, and the impact is stronger in the short term than in the medium and long term; the effect of rising interest rates on technological progress is positive, and the effect intensity is also significant in the short term, but weak in the medium and long term, the effect of price monetary policy on output is stronger under moderate uncertainty. (2) Credit growth can promote output growth, and the regulation effect of credit growth on output growth is mainly reflected in the short term under the TVP-VAR model, the effect of credit growth on technological progress is not significant. Further research using counterfactual analysis shows that the uncertain environment will reduce the effect of credit policy on output growth, but the effect is not significant.

Does Investor Sentiment Affect Clean Energy Stock? Evidence from TVP-VAR-Based Connectedness Approach

We investigated the connectedness of the returns and volatility of clean energy stock, technology stock, crude oil, natural gas, and investor sentiment based on the time-varying parameter vector autoregressive (TVP-VAR) connectedness approach. The empirical results indicate that the average total connectedness is higher in the volatility system than in the return system. The investor sentiment has a weak impact on clean energy stock. Our results show that the dynamic total connectedness across assets in the system varies with time. Furthermore, the dynamic total connectedness increases significantly during financial turmoil. Dynamic total volatility connectedness is more sensitive to financial turmoil. By comparing the connectedness estimated by the TVP-VAR model with the rolling-window VAR model, we find the dynamic total return connectedness of the TVP-VAR model is similar to the estimated results of a 200 day rolling-window VAR model.

Survey Expectations and Learning

In this paper, we evaluate a model that describes real-time inflation data together with the inflation expectations measured by the Survey of Professional Forecasters (SPF). We work with a second-order autoregressive model in which the agents learn over time the intercept and persistence coefficients based on real-time data. To model the process of revisions in real time data, we allow for news and noise disturbances. In contrast to the usual time-varying parameter vector autoregression, we use non-linear Kalman filter techniques to estimate the time-varying coefficients of the underlying inflation process. We identify systematic changes in the persistence of the inflation process and in the long-run expected inflation rate that are implied by the model. The inflation forecasts implied by the model are then compared with the SPF forecasts. As we cannot reject the hypothesis that the SPF forecasts are produced based on our model, we re-estimate the model using Survey nowcasts and forecasts as additional observables. This augmented model does not change the nature and magnitude of the time variation in the coefficients of the autoregressive model, but it does help to reduce the uncertainty in the estimates. Overall, the estimated time-variation confirms our results on the perceived inflation process present in estimated DSGE models with learning (Slobodyan and Wouters, 2012a, 2012b).

Agricultural commodity markets and oil prices: An analysis of the dynamic return and volatility connectedness

We investigate the joint and bivariate return and volatility interdependence between various agricultural commodities and oil price shocks. As an alternative of the Diebold and Yilmaz (2012 and 2014) spillover methodology, this paper proposes the application of the fresh time-varying parameter vector autoregression (TVP-VAR) methodology by Antonakakis and Gabauer (2017) during the sample period between January 7, 2000 and September 17, 2020. In addition, this paper pays special attention to the most relevant periods of economic turbulence among the last 20 years: dotcom bubble, Global Financial Crisis (GFC) and COVID-19 pandemic crisis. About the main results, the directional return and volatility connectedness of oil risk shocks is higher than oil demand shocks and, in turn, higher than oil supply shocks. In addition, the dynamic total return and volatility connectedness changes over time, rising during periods of economic crisis. In general, the net return connectedness considerably increases during the three most important crises. Thus, the differences between transmitters' (Canola and Corn) and receivers’ (Orange Juice, Lean Hog, Sugar and Rubber) agricultural commodity markets are emphasized during the GFC and the COVID-19 pandemic crisis. Finally, the net volatility connectedness measure would not show evidence as clear as the net return connectedness measure.

Are oil-clean energy and high technology stock prices in the same straits? Bubbles speculation and time-varying perspectives

Motivated by the drive to improve the performance and growth of clean energy technology amidst related high-tech innovations, the vulnerability of clean energy and high-tech stock prices to oil shocks is examined, by illustrating the potential bubbles and time-varying interactions among the commodities over the period from January 2004 to December 2017. In this regard, we contribute to the literature in two aspects. First, we analyze an empirically important issue with the SADF (Supremum Augmented Dickey-Fuller) approach for explosive bubbles in oil price, clean energy, and high-tech stock prices. Second, the Markov Chain Monte Carlo (MCMC) approach of the Bayesian time-varying parameter Vector Autoregressions model with stochastic volatility (TVP-SVAR) technique is used to account for time-varying and state dependent interactions between commodities. We found that the time varying behavior of the dependence among clean energy, high technology stocks and oil prices is mainly due to major bubbles identified in the underlying series. We established contrasting evidence between the responses of clean energy and high-tech stocks to oil disruption shocks. Moreover, the stock return volatilities of high technology stocks have no effect on investors’ expectations of clean energy returns across different time horizons. Overall, this study presents significantly relevant policy guideline.

The connectedness between Sukuk and conventional bond markets and the implications for investors

PurposeThe purpose of this paper is to examine the connectedness across a variety of Sukuk and conventional bond indices and the implications for optimal asset allocation for the period January 1, 2010–April 30, 2020.Design/methodology/approachThe data set consists of five major Sukuk (Dow Jones Sukuk, Thompson Reuters BPA Malaysia Sukuk, Indonesia Government Sukuk, S&P MENA Sukuk and Tadawul Sukuk and Bonds Index) and five conventional bond indexes, one for developed (USA) and four for emerging markets (Malaysia, Indonesia, Africa and Qatar). This study investigates the connectedness and volatility spillover effects across the aforementioned indices, by following the Diebold and Yilmaz (2012) approach, based on the time-varying parameter vector autoregressive (TVP-VAR) model. In addition, this paper provides optimal hedge ratios and portfolio weights for investors.FindingsThe empirical results show that Sukuk and conventional bond markets are highly integrated and that total connectedness exhibits sensitivity to exogenous shocks. The Dow Jones and the Malaysian Sukuk indices are the primary shock transmitters to other markets. However, the weak volatility spillovers between the Dow Jones and conventional bonds suggest that opportunities for optimal asset allocation may in fact exist. The highest (lowest) hedging effectiveness can be achieved by taking a short position in Malaysian (Qatarian) bonds.Originality/valueTo the best of the knowledge, this is the largest sample taken into account to investigate the connectedness between Sukuk and conventional bonds.

Effects of monetary policy on the exchange rates: A Time-varying analysis

Employing a time-varying parameter vector autoregression (VAR) model, which is identified by high frequency monetary policy surprises, we examine the effects of monetary policy on the exchange rates. Our results indicate that a contractionary monetary policy shock leads to an appreciation of the exchange rate during both conventional and unconventional monetary policy periods in the US. During the period 2008–2012, when unconventional monetary policy was practiced, monetary policy shocks produced greater negative effects on the exchange rate compared with other periods. We find that the maximal appreciation of the exchange rate occurs within 3–4 months during most of the period, providing favorable evidence to support Dornbusch’s (1976) overshooting hypothesis.

Dynamic integration and transmission channels among interest rates and oil price shocks

This paper examines the short term dynamic integration among oil price shocks and interest rates for the U.S.A, Euro area and twelve Asian economies from August 1999 to January 2018 using a Time-Varying Parameter Vector Autoregression (TVP-VAR) with stochastic volatility. First, we found convincing evidence of time variation in the co-movement of interest rates and oil shocks and that the integration levels were highest towards the 2001 financial crisis whereas there is an evidence of decoupling as shown by notable drop in the level of integration during the 2007–2009 economic and Euro-debt crises. In descending order, Singapore, crude oil, Hong Kong, Philippines and the United States are the net-transmitters of shocks while India, Japan and Vietnam are net-receivers. Results from a sub-sample containing highly industrializing economies in Asia, the United States, Euro area and crude oil market suggest that Singapore, Hong Kong and the United States remained top transmitters of shocks whereas the Euro-area, Taiwan, Korea and crude oil market become net receivers of shocks. Results from the analysis of transmission channels suggest that higher integration for the full sample tend to be driven by increasing levels of external exposure through trade and financial linkages, information asymmetry and political stability while financial crisis reduces the level of integration. Lastly, among the highly industrialized markets, time varying integration is also driven by the degree of external exposure as well as both political and financial stability. We also document important policy implications of our findings.

Time-Varying Impacts of Carbon Price Drivers in the EU ETS: A TVP-VAR Analysis

This study analyzes the impacts of different drivers on the pricing of EU carbon futures in various periods by using the time-varying parameter vector autoregressive (TVP-VAR) model. The results indicate that: (1) The relationships between oil, gas, electricity, stock prices and carbon price have significant time-varying characteristics and those relationships have experienced an inversion in 2016. This might be due to the pressure of achieving the “EU 20-20-20” targets and the signing of the Paris Agreement as well as the fine-tuning of the European Union Emissions Trading Scheme (EU ETS). (2) The impacts of different drivers on carbon price are various. The carbon price is more sensitive to oil, gas, electricity prices as well as the stock price before the inversion in the short-term, while its response to changes in the stock price after the inversion is more obvious in the mid-long term. (3) After the signing of the Paris Agreement in the second quarter of 2016, the carbon price has a greater response to changes in its drivers. The oil price’s impact on carbon price became the most significant one among them.

How does economic policy uncertainty connect with the dynamic spillovers between precious metals and bitcoin markets?

In this paper, we examine the role of United States Economic Policy Uncertainty (US EPU) on the connectedness between bitcoin and precious metals using both dynamic spillover and nonparametric causality-in-quantiles approaches. First, using the time varying parameter vector autoregressions spillover test, we found that there is strong connectedness between the Bitcoin and precious metals markets. Second, BDS test shows that nonlinearity is a very crucial factor to be put into consideration when examining the causal effect of economic policy uncertainty on the interactions between Bitcoin and precious metals markets. Third, the non-parametric causality-in-quantile test confirms that the connectedness between the markets and economic policy uncertainty is stronger mostly around the median and higher quantiles. Thus, the connection between bitcoin and precious metals might not act as a hedge or safe haven against economic policy uncertainties especially around these periods. Our findings have pertinent policy implications for investors and markets analysts.

Spillovers in higher moments and jumps across US stock and strategic commodity markets

This paper examines the dynamics of spillover effects on realized estimators of return distributions across US stock, crude oil and gold markets. Using 5-min data from April 11, 2006 to April 29, 2019, daily realized volatility, realized skewness, realized kurtosis, and jumps are computed. The dynamics of spillovers are revealed using the time-varying parameter vector autoregression (TVP-VAR) model, which reflects the way spillovers are shaped by various crisis periods. The results show that all spillovers seem to intensify during crisis periods. Besides the importance of realized volatility spillovers, spillovers in skewness, kurtosis and jumps are notable, although realized volatility spillovers are relatively stronger. Specifically for realized volatility and kurtosis, the stock index is the main net transmitter, whereas for realized skewness and jumps, oil is the main net transmitter. Gold is a net receiver of all realized higher moments and jumps. The findings suggest the utility of considering spillovers of higher moments and jumps when studying US stock, crude oil, and gold markets, which has implications for portfolio and risk management.

The Dynamic Correlation among Financial Leverage, House Price, and Consumer Expenditure in China

With the help of the time-varying parameter vector autoregression with stochastic volatility (TVP-SV-VAR) model and the Bayesian dynamic conditional correlational autoregressive conditional heteroscedasticity (Bayesian DCC-GARCH) model, this study analyzes the interaction mechanism and dynamic correlation among financial leverage, house price, and consumer expenditure (the survey data are collected from China’s National Bureau of Statistics from January 2000 to December 2019; the data on financial leverage and consumer expenditure are from the Wind economic database, and the price of commercial housing was calculated based on the sales volume and area of commercial housing on the official website of China’s National Bureau of Statistics). Empirical results show that an increase in financial leverage significantly increases house prices and reduces consumer expenditure, that a rise in house prices inhibits financial leverage and weakens consumer expenditure, and that an increase in consumer expenditure raises financial leverage and stimulates a rise in house prices. In addition, house price and consumer expenditure are most relevant, followed by financial leverage and consumer expenditure, and then by financial leverage and house price. Therefore, systematic analysis of dynamic correlation among the three variables has important practical significance for formulating appropriate financial policies to stabilize house prices and promote the growth of consumer expenditures. Specially, financial leverage is an important factor to hold back soaring house prices and shrinking consumer expenditure. Therefore, monetary and macroprudential policies should be used to deal with financial leverage variables in order to achieve a balanced and sustainable development of the macroeconomy in China.

Oil prices, policy uncertainty and travel and leisure stocks in China

By utilizing the time-varying parameter vector autoregressive (TVP-VAR) model, we investigate the time-varying effects of oil price shocks and policy uncertainty on travel and leisure stocks from January 2000 to December 2018 in China. The results suggest that the impacts of oil price shocks on travel and leisure stock returns are mainly positive, which support the market inertia theory. However, the effects of policy uncertainty on travel and leisure stock returns fluctuate between positive and negative. Furthermore, the impacts of oil price shocks and policy uncertainty vary significantly across different sub-industries. For classified travel and leisure stocks, oil price shocks and policy uncertainty both have the greatest influences on scenic spot stock returns. Among classified policy uncertainty, the influences of trade policy uncertainty on travel and leisure stocks are the most significant. • • The impacts of oil price shocks on travel and leisure stock returns are mainly positive. • • The effects of policy uncertainty on travel and leisure stock returns vary between positive and negative. • • The impacts of oil price shocks and policy uncertainty vary significantly across different sub-industries. • • Oil price shocks and policy uncertainty both have the greatest influences on scenic spot stock returns. • • The influences of trade policy uncertainty on travel and leisure stocks are the most significant.

The realized volatility of commodity futures: Interconnectedness and determinants

Using high frequency data and connectedness measures based on a time-varying parameter vector autoregression (TVP-VAR) model, we study dynamic connectedness among the realized volatility of 15 commodity futures (Gold, Heating oil, Light crude oil, Natural gas, Copper, Platinum, Cocoa, Coffee, Corn, Cotton, Orange Juice, Soybean, Soybean meal, Sugar, and Wheat) from September 22, 2008 to May 28, 2020. The results show strong and moderate levels of volatility connectedness among energy and metals and moderate connectedness levels within the group of agricultural commodities. Cross-commodity connectedness can explain a large portion of volatility connectedness in some cases, highlighting the importance of conducting realized volatility connectedness within a model allowing realized volatilities to be endogenously and simultaneously determined. Connectedness is robust to alternative specifications and varies with time. It is mostly driven by macroeconomic variables and uncertainty, including the term spread of interest rates and real economic activity. However, the analysis shows that some of the drivers of connectedness differ between upper and lower quantiles.

A closer look into the global determinants of oil price volatility

Abstract In this paper we investigate global determinants of oil price volatility by employing a time-varying parameter vector autoregressive (TVP-VAR) model. We focus on realised volatility and consider the impact from a set of potential determinants including oil supply, oil demand, oil inventory, financial market uncertainty, financial interbank stress, as well as, financial trends in different currencies. We investigate the impact of these factors on realised volatility utilising monthly data over the period 1990:1–2019:5. Findings show that all factors can be conducive to higher levels of realised oil price volatility particularly in the short run. What can further be noticed, is that the magnitude of the corresponding impulse response functions may differ across time and this could largely be attributed to specific intervals of financial crises and economic recessions. Nevertheless, we show that shocks originating to the financial markets tend to be more important for oil price volatility. Our findings are closely linked to the implications regarding the financialisation of the oil market.

Is Broader Better? A Monetary Approach to Forecasting Economic Activity

This paper investigates whether the use of broader Divisia monetary aggregates improves money’s performance in forecasting economic activity within a time-varying parameter vector autoregressive (TVP-VAR) framework. We evaluate entire predictive densities from several alternative models of US output growth and inflation, each using eight different Divisia monetary aggregates. Using the broadest, M4 aggregate produces out-of-sample forecasts which consistently outperform those based on narrower measures of money, pooling of forecasts from several models, and a large-scale, 143-variable model. Our results show that TVP-VARs with Divisia M4 forecast economic activity more accurately than constant-parameter models with alternative or no measures of money.

Tail Risk Connectedness in the Refined Petroleum Market: A First Look at the Impact of the COVID Pandemic

We use the conditional autoregressive value at risk (CAViaR) model in combination with the time-varying parameter vector autoregressive (TVP-VAR) based connectedness approach to study the systematic tail risk transmission considering two types of crude oil (Brent and WTI) and also four refined petroleum products prices (gasoline, heating oil, jet fuel and propane) from January, 17, 1997 to December 11, 2020. For robustness purposes, we utilise three commonly used VaR probability cut-off points; that is, 10%, 5% and 2.5%. The selected sample period covers the first stage of the recent COVID-19 pandemic crisis. Results show that, both heating oil and kerosene are persistent net transmitters of uncertainty episodes in the network, signifying the importance of liquidity in the relevant markets. In addition, the role of the two measures of crude oil appears to shift around 2009 following developments in the energy market. Overall, findings suggest that, connectedness rises with major crisis episodes and that the recent COVID-19 pandemic appears to have the potential to propel tail risk and exposure to losses to levels akin to those of the Global Financial Crisis of 2007.