Skip to main content
SpringerLink
Log in

Who’s behind the wheel? The role of social and media news in driving the stock–bond correlation

  • Original Research
  • Published:
Review of Quantitative Finance and Accounting Aims and scope Submit manuscript

Abstract

This study investigates the impact of both social and news sentiments indices on the dynamic stock–bond correlation across wavelet-based time-scales over the period 1998–2016. Our results show that the news sentiments namely unemployment, tsunami and sanctions exhibit significant effects during expansion at the shortest time-scale of [2–4] days. These predictors remain significant with reverse signs during recession on the long investment horizon. Yet, the predictability of social media sentiments differs from that of news sentiments with the pattern of reversal in sign also presents for some proxies including windstorm and investment flows. Statistically, our further analysis confirmed the predictability of the sentiments out-of-sample. Excluding the news and social media sentiment effects has also resulted in minimizing the value-at-risk of the (40/60) stock/bond portfolios the most at the investment horizon of [32–64] days during recessions. Our results remain the same after performing some robustness checks.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. Nardo et al. (2016) reviews the literature exploring social media and its role in stock market predictability.

  2. Down-sampling means reducing the sample size to the half at each time-scale. For example, 1000 observations on first time-scale will be 500 on the second time-sale, 250 on time-scale 3 and so forth.

  3. The decomposition of the daily return series data ensures that the variance at short time-scale is more than that at the long-time scale. That helps in explaining the excess volatility in the short run periods resulting from the high uncertainty and the behaviour of short-term investors in the market.

  4. Aloui et al. (2021) extracted the sentiment proxy from the American association of individual investors (AAII) database and performed their analysis using the continuous wavelet approach.

  5. These sentiment indicators are defined in the Appendix A.1. We describe the process of construction our sentiments in Sect. 3.1.

  6. The data are provided by Thomson Reuters Financial and Risk Team as part of TRMI product. TRMI covers a plethora of securities and markets, including more than 12,000 companies, 36 commodities and energy subjects, 187 countries, 62 sovereign markets, 45 currencies, and, since 2009, more than 150 cryptocurrencies. For more details, see Thomson Reuters MarketPsych Indices 2.2 User Guide, 25 May 2019, Document Version 1.0.

  7. Also aiming at categorising sentiment measures, Ding et al. (2019) decomposed Baker et al. (2006)’s sentiment into short and long term components and examined their impacts on the cross section of stock returns.

  8. We dropped out some news (social media) constructed PCs if they have high correlation with the other constructed news (social media) ones. The correlation coefficient of 70 per cent if considered cut off point to either consider or drop out the variables from the further analysis.

  9. For extensive review of the multivariate GACRH extensions, see Bauwens et al. (2006) and Silvennoinen and Teräsvirta (2009).

  10. We also perform the log likelihood ratio test by comparing between the performance of the unrestricted (ADCC) and the restricted (DCC) model. The null hypothesis for the superiority of the DCC model is then rejected at the aggregate level and at all time-scales at 5% significance level. The critical value for the test is 3.84. The results from the test are available upon request.

  11. While we also assume the standardised residuals are normally distributed, following Cappiello et al. (2006), we note that the choice of distribution is not statistically important in the estimation of the conditional variance equation.

  12. To estimate our model, we made the order of lagged correlation equals zero.

  13. The diagonal version assumes that equity-bond correlation can differ across different maturity bonds.

  14. However, controlling for only one lag of the daily correlation can be hardly sufficient. To deal with this issue, we controlled for up to 5 lags (a trading week) and our results remain qualitatively and quantitatively similar. Also, based on the information criteria and the log likelihood, our specification is found to be adequate. We thank the reviewer for addressing this point. The results from this additional analysis are available upon request.

  15. Other related studies also ran their regressions with one of the predictors being the lagged correlation (e.g. Christiansen and Ranaldo 2007 among others). However, we also estimated our model without the lagged correlation and the results remain statistically the same, but the adjusted R became low. These findings are available upon request.

  16. Intercepts were included in the regression estimates but not reported to save the space.

  17. Again, the variation in the significance level mirrors those observed in the news sentiments-based regressions, for the full explanation, refer to pages 33 and 34.

  18. We also perform the rolling window exercise. See Appendix C.

  19. The full VaR estimates on time-scales are available upon request. They are omitted to save the space.

References

  • Aloui C, Shahzad SJH, Hkiri B, Hela BH, Khan MA (2021) On the investors’ sentiments and the Islamic stock-bond interplay across investments’ horizons. Pacific-Basin Finance J 65:101491

    Article  Google Scholar 

  • Andersson M, Krylova E, Vähämaa S (2008) Why does the correlation between stock and bond returns vary over time? Appl Financial Econ 18(2):139–151

    Article  Google Scholar 

  • Asgharian H, Christiansen C, Hou AJ (2015) Macro-finance determinants of the long-run stock–bond correlation: The DCC-MIDAS specification. J Financial Econ 14(3):617–642

    Google Scholar 

  • Aslanidis N, Christiansen C (2014) Quantiles of the realized stock–bond correlation and links to the macroeconomy. J Empirical Finance 28:321–331

    Article  Google Scholar 

  • Azar PD, Lo AW (2016) The wisdom of Twitter crowds: Predicting stock market reactions to FOMC meetings via Twitter feeds. J Portfolio Manage 42(5):123–134

    Article  Google Scholar 

  • Baele L, Bekaert G, Inghelbrecht K (2010) The determinants of stock and bond return comovements. Rev Financial Stud 23(6):2374–2428

    Article  Google Scholar 

  • Baker MP, Wurgler J (2006) Investor sentiment and the cross-section of stock returns. J Finance 61(4):1645–1680

    Article  Google Scholar 

  • Baker M, Wurgler J (2007) Investor sentiment in the stock market. J Econ Perspect 21(2):129–152

    Article  Google Scholar 

  • Baker M, Wurgler J (2012) Comovement and predictability relationships between bonds and the cross-section of stocks. Rev Asset Pricing Stud 2(1):57–87

    Article  Google Scholar 

  • Baker SR, Bloom N, Davis SJ (2016) Measuring Economic Policy Uncertainty. Quarterly J Econ 131(4):1593–1636. https://doi.org/10.1093/qje/qjw024

  • Bauwens L, Laurent S, Rombouts JV (2006) Multivariate GARCH models: a survey. J Appl Econ 21(1):79–109

    Article  Google Scholar 

  • Beber A, Brandt MW, Kavajecz KA (2008) Flight-to-quality or flight-to-liquidity? Evidence from the euro-area bond market. Rev Financial Stud 22(3):925–957

    Article  Google Scholar 

  • Bekaert G, Engstrom E, Xing Y (2009) Risk, uncertainty, and asset prices. J Financ Econ 91(1):59–82

    Article  Google Scholar 

  • Bollen J, Mao H, Zeng X (2011) Twitter mood predicts the stock market. J Comput Sci 2(1):1–8

    Article  Google Scholar 

  • Bollerslev T (1990) Modelling the coherence in short-run nominal exchange rates: a multivariate generalized ARCH model. Rev Econ Stat 72(3):498–505

    Article  Google Scholar 

  • Bukovina J (2016) Social media big data and capital markets—An overview. J Behav Exp Finance 11:18–26

    Article  Google Scholar 

  • Caballero RJ, Krishnamurthy A (2008) Collective risk management in a flight to quality episode. J Finance 63(5):2195–2230

    Article  Google Scholar 

  • Campbell JY, Pflueger C, Viceira LM (2014) Monetary policy drivers of bond and equity risks. Harvard University, Cambridge

    Google Scholar 

  • Cappiello L, Engle RF, Sheppard K (2006) Asymmetric dynamics in the correlations of global equity and bond returns. J Financial Econ 4(4):537–572

    Google Scholar 

  • Cappiello L, Kadareja A, Manganelli S (2010) The impact of the Euro on equity markets. J Financial Quant Anal 45(2):473–502

    Article  Google Scholar 

  • Chahine S, Malhotra NK (2018) Impact of social media strategies on stock price: the case of Twitter. Eur J Mark 52(7/8):1526–1549

    Article  Google Scholar 

  • Chen H, De P, Hu Y, Hwang BH (2014) Wisdom of crowds: the value of stock opinions transmitted through social media. Rev Financial Stud 27(5):1367–1403

    Article  Google Scholar 

  • Chiang TC, Li J, Yang SY (2015) Dynamic stock–bond return correlations and financial market uncertainty. Rev Quant Financ Acc 45(1):59–88

    Article  Google Scholar 

  • Christiansen C, Ranaldo A (2007) Realized bond—stock correlation: Macroeconomic announcement effects. J Fut Mark 27:439–469. https://doi.org/10.1002/fut.20258

    Article  Google Scholar 

  • Christiansen C (2010) Decomposing European bond and equity volatility. Int J Finance Econ 15(2):105–122

    Google Scholar 

  • Conlon T, Cotter J, Gencay R (2018) Long-run wavelet-based correlation for financial time series. Eur J Oper Res 271(2):676–696

    Article  Google Scholar 

  • Connolly RA, Stivers C, Sun L (2007) Commonality in the time-variation of stock–stock and stock–bond return comovements. J Financial Markets 10(2):192–218

    Article  Google Scholar 

  • Connolly R, Stivers C, Sun L (2005) Stock market uncertainty and the stock-bond return relation. J Financial Quant Anal 40(1):161–194

    Article  Google Scholar 

  • David A, Veronesi P (2013) What ties return volatilities to price valuations and fundamentals? J Political Econ 121(4):682–746

    Article  Google Scholar 

  • Ding W, Mazouz K, Wang Q (2019) Investor sentiment and the cross-section of stock returns: new theory and evidence. Rev Quant Financ Acc 53(2):493–525

    Article  Google Scholar 

  • Dunteman G (1994) Factor analysis and related techniques. In: Lewis-Beck M, (eds) Principal components analysis. International Handbooks of Quantitative Applications in the Social Sciences 5. (Sage Publications, London), pp. 157–245

  • Engle R (2002) Dynamic conditional correlation: a simple class of multivariate generalized autoregressive conditional heteroscedasticity models. J Bus Econ Stat 20(3):339–350

    Article  Google Scholar 

  • Fang L, Peress J (2009) Media coverage and the cross-section of stock returns. J Finance 64(5):2023–2052

    Article  Google Scholar 

  • Fang L, Yu H, Huang Y (2018) The role of investor sentiment in the long-term correlation between US stock and bond markets. Int Rev Econ Finance 58:127–139

    Article  Google Scholar 

  • Galagedera D, Maharaj EA (2008) Wavelet timescales and conditional relationship between higher-order systematic co-moments and portfolio returns. Quant Finance 8(2):201–221

    Article  Google Scholar 

  • Gallegati M, Semmler W (2014) Wavelet applications in economics and finance. Springer

    Book  Google Scholar 

  • Gallegati M, Ramsey JB (2013) Bond vs stock market’s Q: Testing for stability across frequencies and over time. J Empirical Finance 24:138–150

    Article  Google Scholar 

  • Garcia D (2013) Sentiment during recessions. J Finance 68(3):1267–1300

    Article  Google Scholar 

  • Gençay R, Selçuk F, Whitcher BJ (2001) An introduction to wavelets and other filtering methods in finance and economics. Elsevier, Amsterdam

    Google Scholar 

  • Glosten LR, Jagannathan R, Runkle DE (1993) On the relation between the expected value and the volatility of the nominal excess return on stocks. J Finance 48(5):1779–1801

    Article  Google Scholar 

  • Gotthelf N, Uhl MW (2019) News sentiment: a new yield curve factor. J Behav Finance 20(1):31–41

    Article  Google Scholar 

  • Gulko L (2002) Decoupling. J Portfolio Manage 28(3):59–66

    Article  Google Scholar 

  • Gupta R, Kollias C, Papadamou S, Wohar ME (2018) News implied volatility and the stock-bond nexus: Evidence from historical data for the USA and the UK markets. J Multinat Financial Manage 47:76–90

    Article  Google Scholar 

  • Huang S, An H, Gao X, Hao X (2016) Unveiling heterogeneities of relations between the entire oil–stock interaction and its components across time scales. Energy Econ 59:70–80

    Article  Google Scholar 

  • Ilmanen A (2003) Stock-bond correlations. J Fixed Income 13(2):55–66

    Article  Google Scholar 

  • Kaiser H (1960) The application of electronic computers to factor analysis. Educ Psychol Measur 20(1):141–151

    Article  Google Scholar 

  • Kim SJ, Moshirian F, Wu E (2006) Evolution of international stock and bond market integration: influence of the European monetary union. J Bank Finance 30(5):1507–1534

    Article  Google Scholar 

  • Kim S, In F (2010) Portfolio allocation and the investment horizon: a multiscaling approach. Quant Finance 10(4):443–453

    Article  Google Scholar 

  • Kothari S, Lewellen J, Warner JB (2006) Stock returns, aggregate earnings surprises, and behavioral finance. J Financ Econ 79(3):537–568

  • Lehkonen H, Heimonen K (2014) Timescale-dependent stock market comovement: BRICs vs. developed markets. J Empirical Finance 28:90–103

    Article  Google Scholar 

  • Li L (2002) Macroeconomic factors and the correlation of stock and bond returns. Working paper, Yale University

  • Li XM, Zhang B, Gao R (2015) Economic policy uncertainty shocks and stock–bond correlations: Evidence from the US market. Econ Lett 132:91–96

    Article  Google Scholar 

  • Liew JKS, Wang GZ (2016) Twitter sentiment and IPO performance: a cross-sectional examination. J Portfolio Manage 42(4):129–135

    Article  Google Scholar 

  • Liew J, Budavari T (2017) The “sixth” factor—a social media factor derived directly from tweet sentiments. J Portfolio Manage 43(3):102–111

    Article  Google Scholar 

  • Loughran T, Mcdonald B (2011) When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. JFinance 66:35–65. https://doi.org/10.1111/j.1540-6261.2010.01625.x

    Article  Google Scholar 

  • Ludvigson SC, Ng S (2007) The empirical risk-return relation: a factor analysis approach. J Financ Econ 83(1):171–222

    Article  Google Scholar 

  • Maghyereh AI, Abdoh H, Awartani B (2019) Connectedness and hedging between gold and Islamic securities: a new evidence from time-frequency domain approaches. Pacific-Basin Finance J 54:13–28

    Article  Google Scholar 

  • Masset P (2008) Analysis of financial time-series using Fourier and wavelet methods. Available at SSRN 1289420.‏

  • Meng X, Huang CH (2019) The time-frequency co-movement of Asian effective exchange rates: a wavelet approach with daily data. North Am J Econ Finance 48:131–148

    Article  Google Scholar 

  • Mensi W, Hkiri B, Al-Yahyaee KH, Kang SH (2018) Analyzing time–frequency co-movements across gold and oil prices with BRICS stock markets: A VaR based on wavelet approach. Int Rev Econ Finance 54:74–102

    Article  Google Scholar 

  • Moat HS, Preis T, Olivola C, Liu C, Chater N (2014) Using Big Data to predict collective behavior in the real world. Behav Brain Sci 37(1):92–93

    Article  Google Scholar 

  • Nardo M, Petracco-Giudici M, Naltsidis M (2016) Walking down wall street with a tablet: a survey of stock market predictions using the web. J Econ Surv 30(2):356–369

    Article  Google Scholar 

  • Perego ER, Vermeulen WN (2016) Macro-economic determinants of European stock and government bond correlations: a tale of two regions. J Empirical Finance 37:214–232

    Article  Google Scholar 

  • Percival DB, Walden AT (2000) Wavelet methods for time series analysis (Cambridge Series in Statistical and Probabilistic Mathematics)

  • Peterson R (2013) Thomson Reuters Marketpsych Indices (TRMI) White Paper. Inside the Mind of the Market

  • Piñeiro-Chousa J, López-Cabarcos MÁ, Pérez-Pico AM, Ribeiro-Navarrete B (2018) Does social network sentiment influence the relationship between the S&P 500 and gold returns? Int Rev Financial Anal 57:57–64

    Article  Google Scholar 

  • Renault T (2017) Intraday online investor sentiment and return patterns in the US stock market. J Bank Finance 84:25–40

    Article  Google Scholar 

  • Sakemoto R (2018) Co-movement between equity and bond markets. Int Rev Econ Finance 53:25–38

    Article  Google Scholar 

  • Skintzi VD (2019) Determinants of stock-bond market comovement in the Eurozone under model uncertainty. Int Rev Financial Anal 61:20–28

    Article  Google Scholar 

  • Silvennoinen A, Teräsvirta T (2009) Multivariate GARCH models Handbook of financial time series. Springer, Berlin, pp 201–229

    Book  Google Scholar 

  • Sprenger TO, Welpe IM (2011) Tweets and peers: defining industry groups and strategic peers based on investor perceptions of stocks on Twitter. Algorithmic Finance 1(1):57–76

    Article  Google Scholar 

  • Sprenger TO, Tumasjan A, Sandner PG, Welpe IM (2014) Tweets and trades: the information content of stock microblogs. Europ Financial Manage 20(5):926–957

    Article  Google Scholar 

  • Sun A, Lachanski M, Fabozzi FJ (2016) Trade the tweet: Social media text mining and sparse matrix factorization for stock market prediction. Int Rev Financial Anal 48:272–281

    Article  Google Scholar 

  • Tetlock PC (2007) Giving content to investor sentiment: the role of media in the stock market. J Finance 62(3):1139–1168

    Article  Google Scholar 

  • Tetlock PC, Saar-Tsechansky M, Macskassy S (2008) More than words: quantifying language to measure firms’ fundamentals. J Finance 63(3):1437–1467

    Article  Google Scholar 

  • Viceira LM (2012) Bond risk, bond return volatility, and the term structure of interest rates. Int J Forecast 28(1):97–117

    Article  Google Scholar 

  • Yang J, Zhou Y, Wang Z (2009) The stock–bond correlation and macroeconomic conditions: One and a half centuries of evidence. J Bank Finance 33(4):670–680

    Article  Google Scholar 

Download references

Acknowledgements

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author information

Authors and Affiliations

  1. School of Management and Logistic Sciences, German Jordanian University, Amman Madaba Street, P.O. Box 35247, Amman, 11180, Jordan

    Mohammad Alomari

  2. Faculty of Economics and Administrative Sciences, Yarmouk University, Irbid, 21163, Jordan

    Abdel Razzaq Al rababa’a, Ghaith El-Nader & Ahmad Alkhataybeh

Authors
  1. Mohammad Alomari
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Abdel Razzaq Al rababa’a
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ghaith El-Nader
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ahmad Alkhataybeh
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Mohammad Alomari.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix A

See Table

Table 11 News and social sentiment indices definitions
Full size table

11.

Appendix B

See Table

Table 12 Selecting the best ADCC-GARCH specification
Full size table

12.

Appendix C

Notes the figure below shows the rolling regression estimates from the equation:

$$\rho \;{\text{SB}}_{{j,t - 1}} = \alpha + \mathop \sum \limits_{PCn = 1}^{n} \beta 1PC_{j,t} + \beta 2\rho SB_{j,t - 1} + e_{t}$$

The analysis is performed on the time-scales 1 and 5 with a window size of 1250 days and one step ahead in the future. Panel A (B) shows the estimates from the regression with the news (social media) predictors. The shaded areas indicate the NBER recession periods. For further information about full notation in the model, see Table 6 (Fig. 2).

Fig. 2
figure 2figure 2figure 2

Rolling regression analysis on time-scales 1 and 5

Full size image

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Alomari, M., Al rababa’a, A., El-Nader, G. et al. Who’s behind the wheel? The role of social and media news in driving the stock–bond correlation. Rev Quant Finan Acc 57, 959–1007 (2021). https://doi.org/10.1007/s11156-021-00967-4

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11156-021-00967-4

Keywords

JEL Classifications

Search

Navigation