Skip to main content
SpringerLink
Log in

Evidence of feedback trading with Markov switching regimes

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

Abstract

Previous research has concluded that the degree of return autocorrelation observed in index returns varies linearly with the volatility of the series, and that feedback traders are at least partly responsible for this phenomenon. Using daily Australian bond and equity market returns, we test this conclusion directly by using a Markov switching model for changing variance that explicitly allows the autocorrelation of returns to vary with the volatility regime. We find evidence that a significant proportion of investors in both the Australian equity and bond markets are positive feedback traders and are responsible for the observed increase in negative autocorrelation in index returns during periods of high and increasing volatility.

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
Fig. 2

Similar content being viewed by others

Notes

  1. Alternative explanations of return autocorrelations have been largely dismissed in the literature: (a) conditional risk premia—Fama (1971) (countered by e.g. Atchison et al. 1987; Conrad and Kaul 1988, 1989; Lo and MacKinlay 1988, 1990; Mech 1993); (b) nonsynchronous trading—Fischer (1966) and Scholes and Williams (1977) (countered by e.g. Atchison et al. 1987; Ogden 1997); and other market microstructure biases—see Cohen et al. (1980), Mech (1993) and Ogden (1997).

  2. Using an alternative approach, Jackson (2003) examines individual investor trades for evidence of systematic aggregation of investment decisions made by ‘irrational’ investors. He finds evidence of a strong pattern consistent with negative feedback trading.

  3. There also exists a body of work that examines feedback trading hypotheses in non-equity settings. For example, futures markets have been investigated by Kodres (1999) and Koutmos (2002). Specifically, Kodres (1999) analysed individual accounts in the S&P 500 index futures, while Koutmos (2002) looked at four international stock index futures contracts (S&P 500; Nikkei; DAX and CAC40). As yet another alternative example, Cohen and Shin (2002) investigated US Treasury notes. In all three cases the feedback trading hypothesis received reasonable support.

  4. The mixture of two or more ‘normal’ distributions will have the effect of making the overall data appear skewed and leptokurtic assuming differing means and variances across the individual distributions.

  5. A number of possible explanations for this type of behaviour have been proposed in the literature, including the presence of technical analysts’, extrapolative expectations, dynamic trading strategies, the liquidation of positions held by traders unable to meet margin calls, or the use of stop loss orders by investors. Evidence of this type of behaviour for both individual investors and institutions can be found in Bange (2000) and Nofsinger and Sias (1999), respectively. Li and Yung (2004), examining the ADR market, rule out a linkage between institutional herding and positive feedback trading.

  6. All applications in this paper work at the aggregate market level. To simplify notation in the remaining equations, the ‘M’ subscript is suppressed.

  7. These tests are reported later in Table 3.

  8. We gratefully acknowledge an anonymous referee for drawing our attention to this potential line of reasoning.

  9. Recently, Nam et al. (2006) explore asymmetry in the mean reversion of short horizon stock returns effectively in the context of this basic null model. Specifically, the asymmetry is captured by a simple dichotomisation between negative and positive returns, thereby representing a very basic from of non-linearity modelling.

  10. Given a probability π of switching between two regimes, the expected duration of a regime is given by duration = (1 − π)−1 days.

  11. Li and Lin (2003) represents a recent application of a three-regime switching model in the GARCH context, using Taiwan stock index data.

  12. To conserve space, details are suppressed.

  13. Smart traders are rational investors whose demand for an asset is adequately described by the CAPM of Sharpe (1964), i.e. a function of volatility, where an increase in volatility reduces the demand for an asset.

References

  • Atchison MD, Butler KC, Simonds RR (1987) Non-synchronous security trading and market index autocorrelation. J Finance 42:111–118

    Article  Google Scholar 

  • Bange MM (2000) Do the portfolios of small investors reflect positive feedback trading? J Financ Quant Anal 35:239–255

    Article  Google Scholar 

  • Berndt ER, Hall BH, Hall RE, Hausman JA (1974) Estimation and inference in nonlinear structural models. Ann Econ Soc Measure 3:653–665

    Google Scholar 

  • Bollerslev T, Chou TY, Kroner KF (1992) ARCH modelling in finance: a review of the theory and empirical evidence. J Econ 52:5–59

    Google Scholar 

  • Cai J (1994) A markov model of switching-regime ARCH. J Bus Econ Stat 12:309–316

    Article  Google Scholar 

  • Campbell J (1987) Stock returns and the term structure. J Financ Econ 18:373–399

    Article  Google Scholar 

  • Cohen B, Shin H (2002) Positive feedback trading under stress: evidence from the US treasury securities market. AFA 2003 Washington, DC Meetings. http://www.ssrn.com/abstract=341920

  • Cohen KJ, Hawawini GA, Maier SF, Schwartz RA, Whitcomb DK (1980) Implications of microstructure theory for empirical research on stock price behaviour. J Finance 35:249–257

    Article  Google Scholar 

  • Conrad J, Kaul G (1988) Time-variation in expected returns. J Bus 61:409–425

    Article  Google Scholar 

  • Conrad J, Kaul G (1989). Mean reversion in short horizon expected returns. Rev Financ Stud 2:225–240

    Article  Google Scholar 

  • Cutler DM, Poterba JM, Summers LH (1991) Speculative dynamics. Rev Econ Stud 58:529–546

    Article  Google Scholar 

  • Durland JM, McCurdy TH (1994) Duration dependent transitions in a Markov model of U.S. GNP growth. J Bus Econ Stat 12:279–288

    Article  Google Scholar 

  • Engle R (1982) Autoregressive conditional heteroskedasticity with estimates of the variance of the United Kingdom inflation. Econometrica 50:987–1007

    Article  Google Scholar 

  • Fama EF (1971) Foundations of finance. Basic Books, New York

  • Fama E, Schwert G (1977) Asset returns and inflation. J Financ Econ 5:115–146

    Article  Google Scholar 

  • Filardo AJ (1994) Business-cycle phases and their transitional dynamics. J Bus Econ Stat 12:299–308

    Article  Google Scholar 

  • Fischer L (1966) Some new stock market indexes. J Bus 39:191–225

    Article  Google Scholar 

  • Garcia R (1998) Asymptotic null distribution of the likelihood ratio test in Markov switching models. Int Econ Rev 39:763–788

    Article  Google Scholar 

  • Ghysels E (1994) On the periodic structure of the business cycle. J Bus Econ Stat 12:289–298

    Article  Google Scholar 

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

    Article  Google Scholar 

  • Goodwin TH (1993) Business-cycle analysis with a Markov-switching model. J Bus Econ Stat 11:331–339

    Article  Google Scholar 

  • Gray SF (1996) Modelling the conditional distribution of interest rates as a regime-switching process. J Financ Econ 42:27–62

    Article  Google Scholar 

  • Hamilton JD (1989) A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica 57:357–384

    Article  Google Scholar 

  • Hamilton JD (1994) Time series analysis. Princeton University Press, Princeton, N.J.

    Google Scholar 

  • Hamilton JD, Susmel R (1994) Autoregressive conditional heteroscedasticity and changes in regime. J Econ 64:307–333

    Google Scholar 

  • Hansen BE (1992) The likelihood ratio test under non-standard conditions: testing the Markov trend model of GNP. J Appl Econ 7:S61–S82

    Article  Google Scholar 

  • Jackson A (2003) The aggregate behaviour of individual investors. http://www.ssrn.com/abstract=536942

  • Kodres L (1999) The existence and impact of destabilizing positive feedback traders: Evidence from the S&P 500 index futures market. unpublished working SSRN paper, http://www.papers.ssrn.com/sol3/papers.cfm?abstract_id=5701

  • Koutmos G (1997) Feedback trading and the autocorrelation pattern of stock returns: further empirical evidence. J Int Money Finance 16:625–636.

    Article  Google Scholar 

  • Koutmos G (2002) Testing for feedback trading in index futures: a dynamic CAPM approach. EFMA 2002 London meetings. http://www.ssrn.com/abstract=314509

  • Li M-Y, Lin H-W (2003) Examining the volatility of Taiwan stock index returns via a three-regime Markov-switching ARCH model. Rev Quant Finance Account 21:123–139

    Article  Google Scholar 

  • Li D, Yung K (2004) Institutional herding in the ADR market. Rev of Quantitative Finance Account 23:5–17

    Article  Google Scholar 

  • Lo AW, MacKinlay AC (1988) Stock market prices do not follow random walks: evidence from a simple specification test. Rev Financ Stud 1:41–66

    Article  Google Scholar 

  • Lo AW, MacKinlay AC (1990) An econometric analysis of nonsynchronous trading. J Econometrics 45:181–211

    Article  Google Scholar 

  • McKenzie MD, Faff RW (2003) The determinants of conditional autocorrelation in stock returns. J Financ Res 26:259–274

    Article  Google Scholar 

  • Mech TS (1993) Portfolio return autocorrelation. J Financ Econ 34:307–344

    Article  Google Scholar 

  • Nam K, Kim S-W, Arize A (2006) Mean reversion of short-horizon stock returns: asymmetry property. Rev Quant Finance Account 26:137–163

    Article  Google Scholar 

  • Nofsinger JR, Sias RW (1999) Herding and feedback trading by institutional and individual investors. J Finance 54:2263–2295

    Article  Google Scholar 

  • Ogden JP (1997) Empirical analyses of three explanations for the positive autocorrelation of short horizon stock index returns. Rev Quant Finance Account 9:203–217

    Article  Google Scholar 

  • Pagan A, Hong YS (1991) Nonparametric estimation and the risk premium. In: Barnett WA, Powell J, Tauchen GE (eds) Semiparametric and nonparametric methods in econometrics and statistics. Cambridge University Press, Cambridge, pp 51–75

    Google Scholar 

  • Schaller H, Van Norden S (1993) The predictability of stock market regimes: evidence from the Toronto stock exchange. Rev Econ Stat 75:505–510

    Article  Google Scholar 

  • Scholes M, Williams JT (1977) Estimating betas from non-synchronous data. J Financ Econ 5:309–327

    Article  Google Scholar 

  • Sentana E, Wadhwani S (1992) Feedback traders and stock return autocorrelations: evidence from a century of daily data. Econ J 102:415–425

    Article  Google Scholar 

  • Sharpe W (1964) Capital market prices: a theory of market equilibrium under conditions of risk. J Finance 19:425–442

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Accounting and Finance, Monash University, P.O. Box 11E, Clayton, VIC, 3800, Australia

    Warren G. Dean & Robert W. Faff

Authors
  1. Warren G. Dean
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Robert W. Faff
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Robert W. Faff.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Dean, W.G., Faff, R.W. Evidence of feedback trading with Markov switching regimes. Rev Quant Finan Acc 30, 133–151 (2008). https://doi.org/10.1007/s11156-007-0047-6

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11156-007-0047-6

Keywords

JEL Classification

Search

Navigation