Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Uncertainty Due to Infectious Diseases and Energy Market Volatility

Uncertainty Due to Infectious Diseases and Energy Market Volatility Salisu, A., & Adediran, I. (2020). Uncertainty Due to Infectious Diseases and Energy Market Volatility. Energy RESEARCH LETTERS, 1(2). https://doi.org/10.46557/001c.14185 COVID-19 and Energy Unc Uncertaint ertainty Due to Inf y Due to Infec ectious Diseases and Energy Mark tious Diseases and Energy Market V et Volatilit olatility y 1 1 Afees Salisu , Idris Adediran Centre for Econometric & Allied Research,, University of Ibadan, Nigeria Keywords: predictability, volatility, pandemics, energy markets https://doi.org/10.46557/001c.14185 Energy RESEARCH LETTERS Vol. 1, Issue 2, 2020 Motivated by the COVID-19 pandemic, we examine the role of uncertainty due to infectious diseases in predicting energy market volatility using the new dataset on Equity Market Volatility-Infectious Diseases (EMV-ID). We find that the new measure of market uncertainty is a good predictor of energy market volatility in both in-sample and out-of-sample tests. These results have implications for portfolio diversification strategies, which we set aside for future research. COVID-19 pandemic. Our paper contributes to the recent 1. Introduc 1. Introduction tion literature by linking pandemics to economic uncertainties resulting from the COVID-19 pandemic (see Bakas & Tri- In this short communication, we experiment with the In- antafyllou, 2020; Ma et al., 2020). fectious Disease Equity Market Volatility (EMV-ID) dataset Our work suggests possible portfolio diversification op- (see Baker et al., 2020) to tease out information on how portunities for investors given that oil market volatility is uncertainties due to pandemics and epidemics (UPE) affect predictable. While investigating this is outside the scope of energy market volatility. This effort is a novel contribution this note, we leave this aspect of research for future studies. to the energy economics literature. Neither the study of en- The rest of the paper is structured as follows. Section 2 has ergy market volatility nor its linkage with uncertainty are a discussion on data and methodology. Results and discus- new. To-date, the literature has been dominated by debate sions appear in Section 3. The final section concludes. on what constitutes a better predictor of energy market volatility in a horserace between the economic policy un- 2. Data and Methodology 2. Data and Methodology certainty (EPU) and the equity market uncertainty (EMU); see Wei et al. (2017), Hailemariam et al. (2019), Liang et Our data cover two variables, namely the energy market al. (2019), and Mei et al. (2019) for studies on EPU and Aloui volatility and a measure of uncertainty (EMV-ID), which we et al. (2016), Baker et al. (2016) and Wen et al. (2019) for refer to as UPE. Both data are obtained from the FRED data- studies on EMU. The COVID-19 pandemic has introduced base of the Federal Reserve Bank of St. Louis. The energy a different dimension to the functioning of energy markets (see Devpura & Narayan, 2020; Huang & Zheng, 2020; market volatility index used in this study was constructed by the CBOE (the Chicago Board Options Exchange) while Narayan, 2020). The present paper contributes to this the EMV-ID was developed by Baker et al. (2020). The for- evolving strand of literature on COVID-19 and energy mar- mer has been empirically tested over time and studies have kets through a study of the role of UPE in shaping energy established a negative correlation between the volatility in- price volatility. The channel of effect can be explained as follows. In our experiment with the EMV-ID data, we expect dex and market returns (see Bouri et al., 2018; Shaikh & Padhi, 2015). The EMV-ID, however, is still relatively recent uncertainty due to infectious diseases to increase uncer- and therefore has not received much empirical applications tainty in the energy market. Since the pandemic will slow- compared to other volatility indexes. We utilize daily data down transport, trade and economic activity across the over the period of 3/21/2011 and 6/4/2020 (including the globe, investors may engage in panic selling of stocks to minimize risks which may heighten uncertainty in the mar- COVID-19 period) based on data availability and the need to have the same start and end dates for the two series. The ket (see Diaz & de Gracia, 2017). This can otherwise be analyses are conducted using both the full sample and the theoretically explained by the theory of irreversible invest- sample covering the COVID-19 pandemic. The COVID-19 ment that sees investors in energy market delay investment period analysis is restricted to 2/10/2020 (when deaths as- (see Bernanke, 1983; Henry, 1974) and buyers delay spend- ing (see Hamilton, 2003) in the face of energy market un- sociated with the pandemic became imminent) to 6/4/2020 (when the first draft of the paper was submitted). certainty. The way pandemics can influence energy price For the empirical analysis, we construct a bivariate pre- volatility is also discussed in Devpura and Narayan (2020) dictive model, where the energy market volatility is the pre- and we refer readers to that paper. We hypothesize that UPE dicted series while the predictor is UPE (see Westerlund & will predict energy market volatility. Using a popular predictability model, we show strong ev- Narayan, 2012, 2015): idence of energy market volatility predictability due to UPE. This predictability is discovered also over the period of the 1 A major attraction of the new dataset lies in its scope as it covers uncertainty due to all the known pandemics for over three decades, thus, making it the most comprehensive data on the subject. Uncertainty Due to Infectious Diseases and Energy Market Volatility T Table 1: P able 1: Preliminar reliminary results y results Statistics Statistics F Full sample ull sample CO COVID-19 sample VID-19 sample Energy UPE Energy UPE Mean Mean 24.142 1.421 62.091 28.349 Std. De Std. Dev v. . 11.180 6.074 27.330 16.114 ADF[I(0)] ADF[I(0)] -4.619** -3.687** GAR GARCH [I(0)] CH [I(0)] -8.550*** -103.61*** AR ARCH(2) CH(2) 172.24*** 390.85*** AR ARCH(4) CH(4) 96.65*** 198.00*** AR ARCH(6) CH(6) 160.52*** 152.73*** Q-stat (2) Q-stat (2) 79.847*** 214.90*** Q-stat (4) Q-stat (4) 86.682*** 237.19*** Q-stat (6) Q-stat (6) 96.237*** 244.93*** 2 2 Q Q -stat (2) -stat (2) 365.99*** 799.56*** 2 2 Q Q -stat (4) -stat (4) 559.34*** 1124.7*** 2 2 Q Q -stat (6) -stat (6) 1224.9*** 1438.8*** P Persistence ersistence - 0.9392*** (0.00719) Endogeneity Endogeneity 0.7238*** (0.0690) Nobs Nobs 2350 85 85 Note: We limit the formal tests to the full sample due to the small COVID-19 sample period. UPE denotes Uncertainty due to Pandemics & Epidemics. The symbols ***, ** & * repre- sent the rejection of the underlying null hypotheses for the formal tests at 1% , 5% & 10% levels of significance, respectively. I(0) indicates that the ADF & GARCH-based unit root tests are conducted at levels. The GARCH-based unit root test is the one proposed by Narayan and Liu (2015) and it is considered an alternative to the Narayan and Popp (2010) test due to the data frequency used in this study (see also Salisu & Adeleke, 2016). tion, to resolve the issue of conditional heteroscedasticity effect, Westerlund and Narayan (2012, 2015) suggest pre- weighting all the data by and estimating the resulting where is the energy volatility based on the CBOE calcu- equation with the OLS. lations; is the market uncertainty index due to infec- tious diseases; is zero mean idiosyncratic error term; and 3. R 3. Results and Discussion esults and Discussion the coefficient measures the relative impact of UPE on energy volatility and we allow a maximum of five lags given We begin the analysis based on the full sample data fol- the 5-day daily data frequency as well as the need to capture lowed by the results from the COVID-19 sample period. The more dynamics in the estimation process. Thus, the under- idea is to see if the UPE can predict energy market volatility lying null hypothesis of no predictability involves a joint regardless of the data sample. Due to space constraints, we (Wald) test that . Note that the original specifica- briefly present the preliminary results of the data. Table 1 shows an increase in both energy market volatility and UPE tion of (1) is given as ; however, to re- during the COVID-19 sample period. This implies a positive solve any concern with the endogeneity bias resulting from relation between UPE and energy market volatility. We al- the correlation between and as well as any poten- so find presence of persistence, endogeneity bias, and con- tial persistence effect, we follow the approach of Lewellen ditional heteroscedasticity. The message is for us to control (2004) and Westerlund and Narayan (2012, 2015). Thus, the these features of the data in our predictive regression mod- parameter is derived as (where el. measures the degree of persistence in and is described In the main analysis, we focus on the predictability re- as the bias adjusted ordinary least squares (OLS) estimator sults and forecast evaluation, both of which constitute the of Lewellen (2004), which corrects for any persistence effect main contributions of this study. Panel A of Table 2 con- in the predictive model. The additional term tains results and shows that UPE has a positive and sig- corrects for any endogeneity bias result- nificant impact on energy market volatility. This finding is ing from the correlation between and as well as any timely in the midst of COVID-19. This finding is consis- inherent unit root problem in the variable. In addi- tent with the literature demonstrating that market volatili- 2 See Westerlund and Narayan (2015, 2015) for computational details. Another major attraction of this technique lies in its ability to isolate the predictor(s) of interest in the estimation and predictability analyses, thus circumventing parameter proliferation. In other words, the technique helps to limit the predictability analyses to the predictor(s) of interest while it also simultaneously resolves any inherent bias (see Westerlund & Narayan, 2012; Westerlund & Narayan, 2015, for the theoretical expositions; and also Narayan & Gupta, 2015; Narayan et al., 2018; Salisu, Raheem, et al., 2019; Salisu, Swaray, et al., 2019, among others, for recent applications). Energy RESEARCH LETTERS 2 Uncertainty Due to Infectious Diseases and Energy Market Volatility T Table 2: P able 2: Predic redictabilit tability and f y and forecast e orecast evaluation results valuation results P Panel A: Predictability results anel A: Predictability results F Full Sample ull Sample CO COVID-19 Sample VID-19 Sample UPE 0.037485*** 0.03590*** (0.00012) (0.0004) [306.6762] [91.6324] Nobs 2350 85 P Panel B: In-Sample forecast e anel B: In-Sample forecast evaluation valuation F Full Sample ull Sample CO COVID-19 Sample VID-19 Sample Model 1 Model 1 Model 2 Model 2 Model 1 Model 1 Model 2 Model 2 RMSE RMSE 0.0820 0.2933 0.2223 0.5476 Clark & W Clark & West est - 0.0043*** - 0.4997*** (0.0015) (0.0817) [2.8430] [6.1138] Nobs Nobs 1205 1205 65 65 P Panel C: Out-of-Sample forecast e anel C: Out-of-Sample forecast evaluation [F valuation [Full-Sample ull-Sample] ] Model 1 Model 1 Model 2 Model 2 h=10 h=10 h=20 h=20 h=30 h=30 h=10 h=10 h=20 h=20 h=30 h=30 RMSE RMSE 0.0820 0.0824 0.0837 0.2928 0.2928 0.2939 Clark & W Clark & West est 0.0040*** 0.0037*** 0.0030*** (0.0015) (0.0015) (0.0015) [2.6556] [2.4267] [1.9758] Nobs Nobs 1215 1225 1235 1215 1225 1235 P Panel D: Out-of-Sample forecast e anel D: Out-of-Sample forecast evaluation [ valuation [CO COVID-Sample VID-Sample] ] Model 1 Model 1 Model 2 Model 2 h=5 h=5 h=10 h=10 h=20 h=20 h=5 h=5 h=10 h=10 h=20 h=20 RMSE RMSE 0.21615 0.2163 0.2034 0.5285 0.5123 0.4918 Clark & W Clark & West est 0.4617*** 0.4324*** 0.4001*** (0.076984) (0.072418) (0.063894) [5.997383] [5.970464] [6.262076] Nobs Nobs 70 75 85 70 75 85 Note: UPE denotes Uncertainty due to Pandemics & Epidemics. The reported statistics in Panel A are obtained from the joint significance test of five lags of the index expressed in natural logs. Model 1 incorporates the UPE predictor while Model 2 is the historical average model. Thus, the former is the unrestricted model while the latter is the restricted model. Nobs denotes number of observations. The asterisks ***, ** & * imply statistical significance at the 1%, 5% & 10% levels, respectively. Values in parentheses - ( ) denote standard er- rors while those reported in square brackets – [ ] are for t-statistics. The results for the Clark & West test are reported for the model under the null (i.e. Model 2). The RMSE reported for Model 1 is a version of Clark & West (2007) which adjusts the difference in mean squared prediction errors to account for the additional predictors in the model. The null hypothe- sis of a zero coefficient is rejected if this statistic is greater than +1.282 (for a one sided 0.10 test), +1.645 (for a one sided 0.05 test) and +2.00 for 0.01 test (for a one sided 0.01 test) (see Clark & West, 2007). ty is due to economic policy and financial market uncertain- series. Since the two models are nested as the historical av- ty (see, Aloui et al., 2016; Baker et al., 2016; Hailemariam erage is a restricted version of Equation (1), their forecast et al., 2019; Liang et al., 2019; Mei et al., 2019; Wei et al., performance comparison can easily be conducted using the 2017; Wen et al., 2019). Clark and West (2007) [CW] test. The way the CW test equa- tion is constructed in this, the rejection of the null hypoth- esis implies superiority of the UPE-based model for ener- 3.1 Forecast e 3.1 Forecast evaluation valuation gy market volatility over the benchmark model. For the full We further evaluate the forecast performance of Equa- sample analyses, we adopt a 50:50 data split for the in-sam- ple and out-of-sample forecast evaluations. Three out-of- tion (1) by comparing its forecast results with those ob- sample forecast horizons, namely a 10-day, a 20-day and tained from a historical average model, which is a typical a 30-day ahead forecast horizons are considered. A rolling (baseline) predictive model for most financial and economic Energy RESEARCH LETTERS 3 Uncertainty Due to Infectious Diseases and Energy Market Volatility regression approach to forecasting is employed consistent 4. C 4. Conclusion onclusion with the forecasting literature (see Salisu, Swaray, et al., 2019; Salisu & Adeleke, 2016). However, for the COVID-19 In this study, we examine the role of uncertainty due sample, we adopt a 75:25 data split due to data limitations. to pandemics and epidemics in the predictability of energy The results are presented in Table 2 (Panel B for the in- market volatility. We utilize the new dataset by Baker et sample forecast and Panels C & D for the out-of-sample al. (2020) and employ a technique that accommodates the forecasts). We find that the model that incorporates UPE salient features of the relevant series. We find that the new outperforms the benchmark model that ignores it regard- uncertainty data is a good predictor of energy market less of the data sample. One clear implication of the results volatility, with significant in-sample and out-of-sample is that investors in the energy market may need to consider forecasting ability. This conclusion complements the this health risk (due to COVID-19) in the valuation of risk- emerging literature on the vulnerability of the energy mar- adjusted returns for energy stocks in particular and perhaps ket to the COVID-19 pandemic and extending the analysis in their diversification of financial assets in general. Further to other commodity markets will enrich the literature. economic significance tests are needed to confirm these claims which we leave for future studies. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC- BY-SA-4.0). View this license’s legal deed at https://creativecommons.org/licenses/by-sa/4.0 and legal code at https://cre- ativecommons.org/licenses/by-sa/4.0/legalcode for more information. Energy RESEARCH LETTERS 4 Uncertainty Due to Infectious Diseases and Energy Market Volatility REFERENCES Aloui, R., Gupta, R., & Miller, S. M. (2016). Henry, C. (1974). Investment decisions under Uncertainty and Crude Oil Returns. Energy uncertainty: The irreversibility effect. American Economics, 55, 92–100. https://doi.org/10.1016/j.ene Economic Review, 974(64), 1006–1012. co.2016.01.012 Huang, W., & Zheng, Y. (2020). COVID-19: Structural Bakas, D., & Triantafyllou, A. (2020). Commodity changes in the relationship between investor price volatility and the economic uncertainty of sentiment and crude oil futures price. Energy pandemics. Economics Letters, 90. https://doi.org/1 Research Letters, 1(2), In press. https://doi.org/10.465 0.1016/j.econlet.2020.109283 57/001c.13685 Baker, S. R., Bloom, N., & Davis, S. J. (2016). Lewellen, J. (2004). Predicting returns with financial Measuring Economic Policy Uncertainty. The ratios. Journal of Financial Economics, 74(2), Quarterly Journal of Economics, 131(4), 1593–1636. h 209–235. https://doi.org/10.1016/j.jfineco.2002.11.00 ttps://doi.org/10.1093/qje/qjw024 2 Baker, S. R., Bloom, N., Davis, S. J., & Terry, S. J. Liang, C., Wei, Y., Li, X., Zhang, X., & Zhang, Y. (2020). COVID-Induced Economic Uncertainty. (2019). Uncertainty and crude oil market volatility: National Bureau of Economic Research. https://doi.or New evidence. Applied Economics, 52(27), g/10.3386/w26983 2945–2959. https://doi.org/10.1080/00036846.2019.1 Bernanke, B. S. (1983). Irreversibility, uncertainty, and cyclical investment. The Quarterly Journal of Ma, C., Rogers, J. H., & Zhou, S. (2020). Global Economics, 98(1), 85–106. https://doi.org/10.2307/18 Economic and Financial Effects of 21st Century 85568 Pandemics and Epidemics. Covid Economics: Vetted and Real-Time Papers, 5, 56–78. https://doi.org/10.21 39/ssrn.3565646 Bouri, E., Gupta, R., Hosseini, S., & Lau, C. K. M. (2018). Does global fear predict fear in BRICS stock markets? Evidence from a Bayesian Graphical Mei, D., Zeng, Q., Cao, X., & Diao, X. (2019). Structural VAR model. Emerging Markets Review, 34, Uncertainty and oil volatility: New evidence. Physica 124–142. https://doi.org/10.1016/j.ememar.2017.11.0 A: Statistical Mechanics and Its Applications, 525, 04 155–163. https://doi.org/10.1016/j.physa.2019.03.043 Clark, T. E., & West, K. D. (2007). Approximately Narayan, P. K. (2020). Oil price news and COVID-19-Is normal tests for equal predictive accuracy in nested there any connection? Energy Research Letters, 1(1). models. Journal of Econometrics, 138(1), 291–311. htt https://doi.org/10.46557/001c.13176 ps://doi.org/10.1016/j.jeconom.2006.05.023 Narayan, P. K., & Gupta, R. (2015). Has oil price Devpura, N., & Narayan, P. K. (2020). Hourly oil price predicted stock returns for over a century? Energy volatility: The role of COVID-19. Energy Research Econ, 48, 18–23. Letters, 1(2), In press. https://doi.org/10.46557/001 c.13683 Narayan, P. K., & Liu, R. (2015). A unit root model for trending time-series energy variables. Energy Diaz, E. M., & de Gracia, F. P. (2017). Oil price shocks Economics, 50, 391–402. https://doi.org/10.1016/j.en and stock returns of oil and gas corporations. Finance eco.2014.11.021 Research Letters, 20, 75–80. https://doi.org/10.1016/ j.frl.2016.09.010 Narayan, P. K., Phan, D. H. B., & Sharma, S. S. (2018). Does Islamic stock sensitivity to oil prices have Hailemariam, A., Smyth, R., & Zhang, X. (2019). Oil economic significance? Pacific-Basin Finance Journal, prices and economic policy uncertainty: Evidence 53, 497–512. https://doi.org/10.1016/j.pacfin.2018.0 from a nonparametric panel data model. Energy 4.003 Economics, 83, 40–51. https://doi.org/10.1016/j.enec o.2019.06.010 Narayan, P. K., & Popp, S. (2010). A new unit root test with two structural breaks in level and slope at Hamilton, J. D. (2003). What is an oil shock? Journal unknown time. Journal of Applied Statistics, 37(9), of Econometrics, 113(2), 363–398. https://doi.org/1 1425–1438. https://doi.org/10.1080/02664760903039 0.1016/s0304-4076(02)00207-5 883 Energy RESEARCH LETTERS 5 Uncertainty Due to Infectious Diseases and Energy Market Volatility Salisu, A. A., & Adeleke, A. I. (2016). Further Wei, Y., Liu, J., Lai, X., & Hu, Y. (2017). Which application of Narayan and Liu (2015) unit root model determinant is the most informative in forecasting for trending time series. Economic Modelling, 55, crude oil market volatility: Fundamental, speculation, 305–314. https://doi.org/10.1016/j.econmod.2016.0 or uncertainty? Energy Economics, 68, 141–150. http 2.026 s://doi.org/10.1016/j.eneco.2017.09.016 Salisu, A. A., Raheem, I. D., & Ndako, U. B. (2019). A Wen, F., Zhao, Y., Zhang, M., & Hu, C. (2019). sectoral analysis of asymmetric nexus between oil Forecasting realized volatility of crude oil futures price and stock returns. International Review of with equity market uncertainty. Applied Economics, Economics & Finance, 61, 241–259. https://doi.org/1 51(59), 6411–6427. https://doi.org/10.1080/0003684 0.1016/j.iref.2019.02.005 6.2019.1619023 Salisu, A. A., Swaray, R., & Oloko, T. F. (2019). Westerlund, J., & Narayan, P. K. (2012). Does the Improving the predictability of the oil-US stock choice of estimator matter when forecasting returns? nexus: The role of macroeconomic variables. Journal of Banking & Finance, 36(9), 2632–2640. http Economic Modelling, 76, 153–171. https://doi.org/1 s://doi.org/10.1016/j.jbankfin.2012.06.005 0.1016/j.econmod.2018.07.029 Westerlund, J., & Narayan, P. K. (2015). Testing for Shaikh, I., & Padhi, P. (2015). The implied volatility Predictability in Conditionally Heteroskedastic Stock index: Is ‘investor fear gauge’ or ‘forward-looking’? Returns. Journal of Financial Econometrics, 13(2), Borsa Istanbul Review, 15(1), 44–52. https://doi.org/1 342–375. https://doi.org/10.1093/jjfinec/nbu001 0.1016/j.bir.2014.10.001 Energy RESEARCH LETTERS http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Energy RESEARCH LETTERS Unpaywall

Uncertainty Due to Infectious Diseases and Energy Market Volatility

Salisu, Afees; Adediran, Idris
Energy RESEARCH LETTERS – Aug 13, 2020
Download PDF
6 pages

Loading next page...
 
/lp/unpaywall/uncertainty-due-to-infectious-diseases-and-energy-market-volatility-pBL0dGmeJe

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
2652-6433
DOI
10.46557/001c.14185
Publisher site
See Article on Publisher Site

Abstract

Salisu, A., & Adediran, I. (2020). Uncertainty Due to Infectious Diseases and Energy Market Volatility. Energy RESEARCH LETTERS, 1(2). https://doi.org/10.46557/001c.14185 COVID-19 and Energy Unc Uncertaint ertainty Due to Inf y Due to Infec ectious Diseases and Energy Mark tious Diseases and Energy Market V et Volatilit olatility y 1 1 Afees Salisu , Idris Adediran Centre for Econometric & Allied Research,, University of Ibadan, Nigeria Keywords: predictability, volatility, pandemics, energy markets https://doi.org/10.46557/001c.14185 Energy RESEARCH LETTERS Vol. 1, Issue 2, 2020 Motivated by the COVID-19 pandemic, we examine the role of uncertainty due to infectious diseases in predicting energy market volatility using the new dataset on Equity Market Volatility-Infectious Diseases (EMV-ID). We find that the new measure of market uncertainty is a good predictor of energy market volatility in both in-sample and out-of-sample tests. These results have implications for portfolio diversification strategies, which we set aside for future research. COVID-19 pandemic. Our paper contributes to the recent 1. Introduc 1. Introduction tion literature by linking pandemics to economic uncertainties resulting from the COVID-19 pandemic (see Bakas & Tri- In this short communication, we experiment with the In- antafyllou, 2020; Ma et al., 2020). fectious Disease Equity Market Volatility (EMV-ID) dataset Our work suggests possible portfolio diversification op- (see Baker et al., 2020) to tease out information on how portunities for investors given that oil market volatility is uncertainties due to pandemics and epidemics (UPE) affect predictable. While investigating this is outside the scope of energy market volatility. This effort is a novel contribution this note, we leave this aspect of research for future studies. to the energy economics literature. Neither the study of en- The rest of the paper is structured as follows. Section 2 has ergy market volatility nor its linkage with uncertainty are a discussion on data and methodology. Results and discus- new. To-date, the literature has been dominated by debate sions appear in Section 3. The final section concludes. on what constitutes a better predictor of energy market volatility in a horserace between the economic policy un- 2. Data and Methodology 2. Data and Methodology certainty (EPU) and the equity market uncertainty (EMU); see Wei et al. (2017), Hailemariam et al. (2019), Liang et Our data cover two variables, namely the energy market al. (2019), and Mei et al. (2019) for studies on EPU and Aloui volatility and a measure of uncertainty (EMV-ID), which we et al. (2016), Baker et al. (2016) and Wen et al. (2019) for refer to as UPE. Both data are obtained from the FRED data- studies on EMU. The COVID-19 pandemic has introduced base of the Federal Reserve Bank of St. Louis. The energy a different dimension to the functioning of energy markets (see Devpura & Narayan, 2020; Huang & Zheng, 2020; market volatility index used in this study was constructed by the CBOE (the Chicago Board Options Exchange) while Narayan, 2020). The present paper contributes to this the EMV-ID was developed by Baker et al. (2020). The for- evolving strand of literature on COVID-19 and energy mar- mer has been empirically tested over time and studies have kets through a study of the role of UPE in shaping energy established a negative correlation between the volatility in- price volatility. The channel of effect can be explained as follows. In our experiment with the EMV-ID data, we expect dex and market returns (see Bouri et al., 2018; Shaikh & Padhi, 2015). The EMV-ID, however, is still relatively recent uncertainty due to infectious diseases to increase uncer- and therefore has not received much empirical applications tainty in the energy market. Since the pandemic will slow- compared to other volatility indexes. We utilize daily data down transport, trade and economic activity across the over the period of 3/21/2011 and 6/4/2020 (including the globe, investors may engage in panic selling of stocks to minimize risks which may heighten uncertainty in the mar- COVID-19 period) based on data availability and the need to have the same start and end dates for the two series. The ket (see Diaz & de Gracia, 2017). This can otherwise be analyses are conducted using both the full sample and the theoretically explained by the theory of irreversible invest- sample covering the COVID-19 pandemic. The COVID-19 ment that sees investors in energy market delay investment period analysis is restricted to 2/10/2020 (when deaths as- (see Bernanke, 1983; Henry, 1974) and buyers delay spend- ing (see Hamilton, 2003) in the face of energy market un- sociated with the pandemic became imminent) to 6/4/2020 (when the first draft of the paper was submitted). certainty. The way pandemics can influence energy price For the empirical analysis, we construct a bivariate pre- volatility is also discussed in Devpura and Narayan (2020) dictive model, where the energy market volatility is the pre- and we refer readers to that paper. We hypothesize that UPE dicted series while the predictor is UPE (see Westerlund & will predict energy market volatility. Using a popular predictability model, we show strong ev- Narayan, 2012, 2015): idence of energy market volatility predictability due to UPE. This predictability is discovered also over the period of the 1 A major attraction of the new dataset lies in its scope as it covers uncertainty due to all the known pandemics for over three decades, thus, making it the most comprehensive data on the subject. Uncertainty Due to Infectious Diseases and Energy Market Volatility T Table 1: P able 1: Preliminar reliminary results y results Statistics Statistics F Full sample ull sample CO COVID-19 sample VID-19 sample Energy UPE Energy UPE Mean Mean 24.142 1.421 62.091 28.349 Std. De Std. Dev v. . 11.180 6.074 27.330 16.114 ADF[I(0)] ADF[I(0)] -4.619** -3.687** GAR GARCH [I(0)] CH [I(0)] -8.550*** -103.61*** AR ARCH(2) CH(2) 172.24*** 390.85*** AR ARCH(4) CH(4) 96.65*** 198.00*** AR ARCH(6) CH(6) 160.52*** 152.73*** Q-stat (2) Q-stat (2) 79.847*** 214.90*** Q-stat (4) Q-stat (4) 86.682*** 237.19*** Q-stat (6) Q-stat (6) 96.237*** 244.93*** 2 2 Q Q -stat (2) -stat (2) 365.99*** 799.56*** 2 2 Q Q -stat (4) -stat (4) 559.34*** 1124.7*** 2 2 Q Q -stat (6) -stat (6) 1224.9*** 1438.8*** P Persistence ersistence - 0.9392*** (0.00719) Endogeneity Endogeneity 0.7238*** (0.0690) Nobs Nobs 2350 85 85 Note: We limit the formal tests to the full sample due to the small COVID-19 sample period. UPE denotes Uncertainty due to Pandemics & Epidemics. The symbols ***, ** & * repre- sent the rejection of the underlying null hypotheses for the formal tests at 1% , 5% & 10% levels of significance, respectively. I(0) indicates that the ADF & GARCH-based unit root tests are conducted at levels. The GARCH-based unit root test is the one proposed by Narayan and Liu (2015) and it is considered an alternative to the Narayan and Popp (2010) test due to the data frequency used in this study (see also Salisu & Adeleke, 2016). tion, to resolve the issue of conditional heteroscedasticity effect, Westerlund and Narayan (2012, 2015) suggest pre- weighting all the data by and estimating the resulting where is the energy volatility based on the CBOE calcu- equation with the OLS. lations; is the market uncertainty index due to infec- tious diseases; is zero mean idiosyncratic error term; and 3. R 3. Results and Discussion esults and Discussion the coefficient measures the relative impact of UPE on energy volatility and we allow a maximum of five lags given We begin the analysis based on the full sample data fol- the 5-day daily data frequency as well as the need to capture lowed by the results from the COVID-19 sample period. The more dynamics in the estimation process. Thus, the under- idea is to see if the UPE can predict energy market volatility lying null hypothesis of no predictability involves a joint regardless of the data sample. Due to space constraints, we (Wald) test that . Note that the original specifica- briefly present the preliminary results of the data. Table 1 shows an increase in both energy market volatility and UPE tion of (1) is given as ; however, to re- during the COVID-19 sample period. This implies a positive solve any concern with the endogeneity bias resulting from relation between UPE and energy market volatility. We al- the correlation between and as well as any poten- so find presence of persistence, endogeneity bias, and con- tial persistence effect, we follow the approach of Lewellen ditional heteroscedasticity. The message is for us to control (2004) and Westerlund and Narayan (2012, 2015). Thus, the these features of the data in our predictive regression mod- parameter is derived as (where el. measures the degree of persistence in and is described In the main analysis, we focus on the predictability re- as the bias adjusted ordinary least squares (OLS) estimator sults and forecast evaluation, both of which constitute the of Lewellen (2004), which corrects for any persistence effect main contributions of this study. Panel A of Table 2 con- in the predictive model. The additional term tains results and shows that UPE has a positive and sig- corrects for any endogeneity bias result- nificant impact on energy market volatility. This finding is ing from the correlation between and as well as any timely in the midst of COVID-19. This finding is consis- inherent unit root problem in the variable. In addi- tent with the literature demonstrating that market volatili- 2 See Westerlund and Narayan (2015, 2015) for computational details. Another major attraction of this technique lies in its ability to isolate the predictor(s) of interest in the estimation and predictability analyses, thus circumventing parameter proliferation. In other words, the technique helps to limit the predictability analyses to the predictor(s) of interest while it also simultaneously resolves any inherent bias (see Westerlund & Narayan, 2012; Westerlund & Narayan, 2015, for the theoretical expositions; and also Narayan & Gupta, 2015; Narayan et al., 2018; Salisu, Raheem, et al., 2019; Salisu, Swaray, et al., 2019, among others, for recent applications). Energy RESEARCH LETTERS 2 Uncertainty Due to Infectious Diseases and Energy Market Volatility T Table 2: P able 2: Predic redictabilit tability and f y and forecast e orecast evaluation results valuation results P Panel A: Predictability results anel A: Predictability results F Full Sample ull Sample CO COVID-19 Sample VID-19 Sample UPE 0.037485*** 0.03590*** (0.00012) (0.0004) [306.6762] [91.6324] Nobs 2350 85 P Panel B: In-Sample forecast e anel B: In-Sample forecast evaluation valuation F Full Sample ull Sample CO COVID-19 Sample VID-19 Sample Model 1 Model 1 Model 2 Model 2 Model 1 Model 1 Model 2 Model 2 RMSE RMSE 0.0820 0.2933 0.2223 0.5476 Clark & W Clark & West est - 0.0043*** - 0.4997*** (0.0015) (0.0817) [2.8430] [6.1138] Nobs Nobs 1205 1205 65 65 P Panel C: Out-of-Sample forecast e anel C: Out-of-Sample forecast evaluation [F valuation [Full-Sample ull-Sample] ] Model 1 Model 1 Model 2 Model 2 h=10 h=10 h=20 h=20 h=30 h=30 h=10 h=10 h=20 h=20 h=30 h=30 RMSE RMSE 0.0820 0.0824 0.0837 0.2928 0.2928 0.2939 Clark & W Clark & West est 0.0040*** 0.0037*** 0.0030*** (0.0015) (0.0015) (0.0015) [2.6556] [2.4267] [1.9758] Nobs Nobs 1215 1225 1235 1215 1225 1235 P Panel D: Out-of-Sample forecast e anel D: Out-of-Sample forecast evaluation [ valuation [CO COVID-Sample VID-Sample] ] Model 1 Model 1 Model 2 Model 2 h=5 h=5 h=10 h=10 h=20 h=20 h=5 h=5 h=10 h=10 h=20 h=20 RMSE RMSE 0.21615 0.2163 0.2034 0.5285 0.5123 0.4918 Clark & W Clark & West est 0.4617*** 0.4324*** 0.4001*** (0.076984) (0.072418) (0.063894) [5.997383] [5.970464] [6.262076] Nobs Nobs 70 75 85 70 75 85 Note: UPE denotes Uncertainty due to Pandemics & Epidemics. The reported statistics in Panel A are obtained from the joint significance test of five lags of the index expressed in natural logs. Model 1 incorporates the UPE predictor while Model 2 is the historical average model. Thus, the former is the unrestricted model while the latter is the restricted model. Nobs denotes number of observations. The asterisks ***, ** & * imply statistical significance at the 1%, 5% & 10% levels, respectively. Values in parentheses - ( ) denote standard er- rors while those reported in square brackets – [ ] are for t-statistics. The results for the Clark & West test are reported for the model under the null (i.e. Model 2). The RMSE reported for Model 1 is a version of Clark & West (2007) which adjusts the difference in mean squared prediction errors to account for the additional predictors in the model. The null hypothe- sis of a zero coefficient is rejected if this statistic is greater than +1.282 (for a one sided 0.10 test), +1.645 (for a one sided 0.05 test) and +2.00 for 0.01 test (for a one sided 0.01 test) (see Clark & West, 2007). ty is due to economic policy and financial market uncertain- series. Since the two models are nested as the historical av- ty (see, Aloui et al., 2016; Baker et al., 2016; Hailemariam erage is a restricted version of Equation (1), their forecast et al., 2019; Liang et al., 2019; Mei et al., 2019; Wei et al., performance comparison can easily be conducted using the 2017; Wen et al., 2019). Clark and West (2007) [CW] test. The way the CW test equa- tion is constructed in this, the rejection of the null hypoth- esis implies superiority of the UPE-based model for ener- 3.1 Forecast e 3.1 Forecast evaluation valuation gy market volatility over the benchmark model. For the full We further evaluate the forecast performance of Equa- sample analyses, we adopt a 50:50 data split for the in-sam- ple and out-of-sample forecast evaluations. Three out-of- tion (1) by comparing its forecast results with those ob- sample forecast horizons, namely a 10-day, a 20-day and tained from a historical average model, which is a typical a 30-day ahead forecast horizons are considered. A rolling (baseline) predictive model for most financial and economic Energy RESEARCH LETTERS 3 Uncertainty Due to Infectious Diseases and Energy Market Volatility regression approach to forecasting is employed consistent 4. C 4. Conclusion onclusion with the forecasting literature (see Salisu, Swaray, et al., 2019; Salisu & Adeleke, 2016). However, for the COVID-19 In this study, we examine the role of uncertainty due sample, we adopt a 75:25 data split due to data limitations. to pandemics and epidemics in the predictability of energy The results are presented in Table 2 (Panel B for the in- market volatility. We utilize the new dataset by Baker et sample forecast and Panels C & D for the out-of-sample al. (2020) and employ a technique that accommodates the forecasts). We find that the model that incorporates UPE salient features of the relevant series. We find that the new outperforms the benchmark model that ignores it regard- uncertainty data is a good predictor of energy market less of the data sample. One clear implication of the results volatility, with significant in-sample and out-of-sample is that investors in the energy market may need to consider forecasting ability. This conclusion complements the this health risk (due to COVID-19) in the valuation of risk- emerging literature on the vulnerability of the energy mar- adjusted returns for energy stocks in particular and perhaps ket to the COVID-19 pandemic and extending the analysis in their diversification of financial assets in general. Further to other commodity markets will enrich the literature. economic significance tests are needed to confirm these claims which we leave for future studies. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC- BY-SA-4.0). View this license’s legal deed at https://creativecommons.org/licenses/by-sa/4.0 and legal code at https://cre- ativecommons.org/licenses/by-sa/4.0/legalcode for more information. Energy RESEARCH LETTERS 4 Uncertainty Due to Infectious Diseases and Energy Market Volatility REFERENCES Aloui, R., Gupta, R., & Miller, S. M. (2016). Henry, C. (1974). Investment decisions under Uncertainty and Crude Oil Returns. Energy uncertainty: The irreversibility effect. American Economics, 55, 92–100. https://doi.org/10.1016/j.ene Economic Review, 974(64), 1006–1012. co.2016.01.012 Huang, W., & Zheng, Y. (2020). COVID-19: Structural Bakas, D., & Triantafyllou, A. (2020). Commodity changes in the relationship between investor price volatility and the economic uncertainty of sentiment and crude oil futures price. Energy pandemics. Economics Letters, 90. https://doi.org/1 Research Letters, 1(2), In press. https://doi.org/10.465 0.1016/j.econlet.2020.109283 57/001c.13685 Baker, S. R., Bloom, N., & Davis, S. J. (2016). Lewellen, J. (2004). Predicting returns with financial Measuring Economic Policy Uncertainty. The ratios. Journal of Financial Economics, 74(2), Quarterly Journal of Economics, 131(4), 1593–1636. h 209–235. https://doi.org/10.1016/j.jfineco.2002.11.00 ttps://doi.org/10.1093/qje/qjw024 2 Baker, S. R., Bloom, N., Davis, S. J., & Terry, S. J. Liang, C., Wei, Y., Li, X., Zhang, X., & Zhang, Y. (2020). COVID-Induced Economic Uncertainty. (2019). Uncertainty and crude oil market volatility: National Bureau of Economic Research. https://doi.or New evidence. Applied Economics, 52(27), g/10.3386/w26983 2945–2959. https://doi.org/10.1080/00036846.2019.1 Bernanke, B. S. (1983). Irreversibility, uncertainty, and cyclical investment. The Quarterly Journal of Ma, C., Rogers, J. H., & Zhou, S. (2020). Global Economics, 98(1), 85–106. https://doi.org/10.2307/18 Economic and Financial Effects of 21st Century 85568 Pandemics and Epidemics. Covid Economics: Vetted and Real-Time Papers, 5, 56–78. https://doi.org/10.21 39/ssrn.3565646 Bouri, E., Gupta, R., Hosseini, S., & Lau, C. K. M. (2018). Does global fear predict fear in BRICS stock markets? Evidence from a Bayesian Graphical Mei, D., Zeng, Q., Cao, X., & Diao, X. (2019). Structural VAR model. Emerging Markets Review, 34, Uncertainty and oil volatility: New evidence. Physica 124–142. https://doi.org/10.1016/j.ememar.2017.11.0 A: Statistical Mechanics and Its Applications, 525, 04 155–163. https://doi.org/10.1016/j.physa.2019.03.043 Clark, T. E., & West, K. D. (2007). Approximately Narayan, P. K. (2020). Oil price news and COVID-19-Is normal tests for equal predictive accuracy in nested there any connection? Energy Research Letters, 1(1). models. Journal of Econometrics, 138(1), 291–311. htt https://doi.org/10.46557/001c.13176 ps://doi.org/10.1016/j.jeconom.2006.05.023 Narayan, P. K., & Gupta, R. (2015). Has oil price Devpura, N., & Narayan, P. K. (2020). Hourly oil price predicted stock returns for over a century? Energy volatility: The role of COVID-19. Energy Research Econ, 48, 18–23. Letters, 1(2), In press. https://doi.org/10.46557/001 c.13683 Narayan, P. K., & Liu, R. (2015). A unit root model for trending time-series energy variables. Energy Diaz, E. M., & de Gracia, F. P. (2017). Oil price shocks Economics, 50, 391–402. https://doi.org/10.1016/j.en and stock returns of oil and gas corporations. Finance eco.2014.11.021 Research Letters, 20, 75–80. https://doi.org/10.1016/ j.frl.2016.09.010 Narayan, P. K., Phan, D. H. B., & Sharma, S. S. (2018). Does Islamic stock sensitivity to oil prices have Hailemariam, A., Smyth, R., & Zhang, X. (2019). Oil economic significance? Pacific-Basin Finance Journal, prices and economic policy uncertainty: Evidence 53, 497–512. https://doi.org/10.1016/j.pacfin.2018.0 from a nonparametric panel data model. Energy 4.003 Economics, 83, 40–51. https://doi.org/10.1016/j.enec o.2019.06.010 Narayan, P. K., & Popp, S. (2010). A new unit root test with two structural breaks in level and slope at Hamilton, J. D. (2003). What is an oil shock? Journal unknown time. Journal of Applied Statistics, 37(9), of Econometrics, 113(2), 363–398. https://doi.org/1 1425–1438. https://doi.org/10.1080/02664760903039 0.1016/s0304-4076(02)00207-5 883 Energy RESEARCH LETTERS 5 Uncertainty Due to Infectious Diseases and Energy Market Volatility Salisu, A. A., & Adeleke, A. I. (2016). Further Wei, Y., Liu, J., Lai, X., & Hu, Y. (2017). Which application of Narayan and Liu (2015) unit root model determinant is the most informative in forecasting for trending time series. Economic Modelling, 55, crude oil market volatility: Fundamental, speculation, 305–314. https://doi.org/10.1016/j.econmod.2016.0 or uncertainty? Energy Economics, 68, 141–150. http 2.026 s://doi.org/10.1016/j.eneco.2017.09.016 Salisu, A. A., Raheem, I. D., & Ndako, U. B. (2019). A Wen, F., Zhao, Y., Zhang, M., & Hu, C. (2019). sectoral analysis of asymmetric nexus between oil Forecasting realized volatility of crude oil futures price and stock returns. International Review of with equity market uncertainty. Applied Economics, Economics & Finance, 61, 241–259. https://doi.org/1 51(59), 6411–6427. https://doi.org/10.1080/0003684 0.1016/j.iref.2019.02.005 6.2019.1619023 Salisu, A. A., Swaray, R., & Oloko, T. F. (2019). Westerlund, J., & Narayan, P. K. (2012). Does the Improving the predictability of the oil-US stock choice of estimator matter when forecasting returns? nexus: The role of macroeconomic variables. Journal of Banking & Finance, 36(9), 2632–2640. http Economic Modelling, 76, 153–171. https://doi.org/1 s://doi.org/10.1016/j.jbankfin.2012.06.005 0.1016/j.econmod.2018.07.029 Westerlund, J., & Narayan, P. K. (2015). Testing for Shaikh, I., & Padhi, P. (2015). The implied volatility Predictability in Conditionally Heteroskedastic Stock index: Is ‘investor fear gauge’ or ‘forward-looking’? Returns. Journal of Financial Econometrics, 13(2), Borsa Istanbul Review, 15(1), 44–52. https://doi.org/1 342–375. https://doi.org/10.1093/jjfinec/nbu001 0.1016/j.bir.2014.10.001 Energy RESEARCH LETTERS

Journal

Energy RESEARCH LETTERS – Unpaywall

Published: Aug 13, 2020

There are no references for this article.