Evidence Of Mean Reversion Research Articles

Are Net Discount Ratios Stationary? Evidence of Mean-Reversion and Persistence

This article extends the debate initiated by Haslag, Nieswiadomy, and Slottje (1991, 1994) and Gamber and Sorensen (1994) in this journal about whether the net discount ratio can be described as a stationary process. Haslag, Nieswiadomy, and Slottje found discount ratios to be stationary. Gamber and Sorensen concluded that they are nonstationary; however, they identified the source of the nonstationarity as a single shift in the mean of the series. Using the Cochrane variance ratio and Campbell-Mankiw decomposition tests, the authors find that the net discount ratio follows a trend stationary process. However, to determine the degree or level of persistence, Lo's Modified R/S Analysis (1991) is used. The authors find that the relative importance of any mean shift is a function of the duration of the discount period for expected earnings.

Are forward premia mean reverting?

The return regression methodology is used to test for mean reversion in the forward market for US T-bills over the period 1964 to 1995. Substantial evidence of mean reversion is found in one- to ten-month forward spreads over the 12 to 24 month horizon. Such evidence is indicative of market inefficiency or speculative dynamics in models with time-invariant term premia. This is not necessarily the case, however, in models with time-varying term premia. We show that forward premia estimated using a multi-factor GARCH model accounts for this evidence, thus reconciling the evidence of mean reversion with market efficiency.

Mean Reversion on the JSE: 1961-1997

The existence of mean reversion in share markets, if demonstrated, would cast grave doubts on the efficiency of such markets. In an earlier article in this journal, Bradfield and Arrington presented evidence of mean reversion on the Johannesburg Stock Exchange. The present paper examines those findings, as well as available data for the period 1996 - 1997, and concludes there is no evidence of mean reversion but finds some modest evidence of mean aversion or ’trending’.

An introduction to security returns

The equity premium - the difference between the return achievable from investment in the equity market (RM ) and the risk-free rate of return (RF )- plays an important part in corporate finance. The expression equity premium (sometimes referred to as the equity risk premium) is used to denote the ex ante expectation of investors. The term excess return refers to the ex post achievement of stock returns over and above the risk-free return. If we compare US and UK returns, we find that total returns, real returns and the value of (RM - RF ) are all marginally higher for the UK. Summarized evidence appears in Table 1 and Table 6. Such greater returns may be due to an increased risk premium related to increasing unexpected inflation. Particularly important in estimating the equity risk premium is whether excess returns are measured using a geometric or an arithmetic mean return. To a significant extent, this question revolves around mean reversion in stock returns. Evidence of mean reversion is substantial, although it cannot be proved unequivocally. Given the weight of evidence of mean reversion, there may be a strong case for the use of a geometric mean with an equity premium of between 3% and 5% - or even less.

Testing for mean reversion in heteroskedastic data II: Autoregression tests based on Gibbs-sampling-augmented randomization

A decade ago Fama and French [Fama, E.G., French, K.R., 1988. Permanent and temporary components of stock prices. J. Political Econ. 96 (2) 246–273] estimated that 40% of variation in stock returns was predictable over horizons of 3–5 yr, which they attributed to a mean reverting stationary component in prices. While it has been clear that the Depression and war years exert a strong influence on these estimates, it has not been clear whether the large returns of that period contribute to the information in the data or rather are a source of noise to be discounted in estimation. This paper uses the Gibbs-sampling-augmented randomization methodology to address the problem of heteroskedasticity in estimation of multi-period return autoregressions. Extending the sample period to 1995, we find little evidence of mean reversion. Examining subsamples, only 1926–1946 provides any evidence of mean reversion, while the post war period is characterized by mean aversion. A test of structural change suggests that this difference between pre and post war periods is significant.

Unraveling the Closed-End Country Fund Discount Puzzle

The authors examine the behavior of closed-end country fund discounts. Their results indicate that the level of an equally weighted index of country fund discounts and returns can be partially explained by measures of investor sentiment as well as the slope of the Treasury yield curve. The authors’ results also suggest that different factors cause developed and emerging market closed-end fund discounts to change. Finally, they present evidence of mean reversion in country fund prices relative to net asset value when a fund is underpriced but find no evidence of mean reversion for overpriced funds.

On the mean-reverting properties of target zone exchange rates: Some evidence from the ERM

One key prediction of the target zone model is that the exchange rate should be mean-reverting within the band. Although this property is widely referred to in the target zone literature, little work has been done to test it. In this paper we seek to remedy this important lacuna in the literature by examining the time-series properties of the seven currencies participating in the Exchange Rate Mechanism (ERM) of the European Monetary System, for the period 13 March 1979 to 9 April 1992. Using standard univariate unit root tests we find little evidence of mean reversion. However, using variance ratio test statistics, which control for heterogeneity in the distribution of exchange rate behaviour, we find much more evidence of mean reversion. Another part of our work includes implementing a panel unit-root test.

A Note on the Riskiness of Long Term Investment on the JSE

This note focuses on the risk of longer term investment in equity on the JSE. Graphical and quantitative evidence of serial dependence in annual returns is presented. Based of 16 years of price data on the JSE the study finds significant evidence of mean reversion in annual returns. This note argues that, due to this serial dependence, the risk of longer term investment in equity, as measured by variance, is probably less than perceived.

Testing for Mean Reversion in Heteroskedastic Data II: Autoregression Tests Based on Gibbs-Sampling-Augmented Randomization

A decade ago Fama and French (1988) estimated that 40% of variation in stock returns was predictable over horizons of 3-5 years, which they attributed to a mean reverting stationary component in prices. While it has been clear that the Depression and war years exert a strong influence on these estimates, it has not been clear whether the large returns of that period contribute to the information in the data or rather are a source of noise to be discounted in estimation. This paper uses the Gibbs-sampling-augmented randomization methodology to address the problem of heteroskedasticity in estimation of multi-period return autoregressions. Extending the sample period to 1995, we find little evidence of mean reversion. Examining subsamples, only 1926-46 provides any evidence of mean reversion, while the post war period is characterized by mean aversion. A test of structural change suggests that this difference between pre and post war periods is significant.

ARE LONG‐HORIZON STOCK RETURNS PREDICTABLE? A BOOTSTRAP ANALYSIS

This paper examines empirical evidence of predictability of long‐horizon real and excess stock returns in the UK using univariate as well as multivariate Variance Ratio tests. In order to estimate the sampling distribution of the test statistics, artificial histories ofstock returns are generated from their empirical distribution using the bootstrap method. This allows the construction of significance levels of the test statistic which are free from distributional assumptions. The empirical results indicate that there is no evidence of mean reversion in stock prices even if a wider information set to forecast stock returns is used and that the significance of historical Variance Ratio statistics has been overstated by previous studies.

Estimation of the Near Unit Root Model of Real Exchange Rates

The time-series properties of real exchange rates, on a number of definitions, for 22 industrial countries during 1979-95 were used to re-examine whether PPP holds. It is shown that if real exchange rates reverted to a constant mean slowly, say by five percent a month, then at standard levels of significance we should expect 11 of the 22 series examined to yield evidence of mean reversion and to reject that hypothesis of a unit root. Using models that imply a constant unconditional mean or trend-stationary productivity changes, we find that only one of the 22 real exchange rates shows evidence against unit roots. This low rate of rejection of unit roots in real exchange rates can be construed as evidence against PPP.

Finance constraints and asset pricing: Evidence on mean reversion

This paper investigates the potential effect of finance constraints on asset pricing, focusing on cross-sectional patterns of mean reversion. Theoretical work by Brock and LeBaron (1990) shows that in a production-economy version of the Lucas asset pricing model, finance constraints can accentuate mean reversion. Using panel data and distinctive Canadian institutional features which allow us to identify finance-constrained firms, we find that constrained firms show more evidence of mean reversion. Using a bootstrapping approach to inference, we show that the differences between constrained and unconstrained firms are statistically significant.

Rational expectations and the demand for money: A nonparametric approach

This paper tests the U.S. demand for money for evidence of the effect of rational expectations of the income and interest rate variables that enter as arguments into that function. The data employed are simple-sum and Divisia aggregates, and the nonparametric tests are of the identification and information orthogonality of the various monetary measures. The Akaike Criterion is used to distinguish among the alternative specifications. While non-rationality is the typical result, Divisia aggregates appear to be more “rational” than simple sum. There is evidence of mean-reversion in interest rates as well.