Predictive regression with order-p autoregressive predictors

Studies of predictive regressions analyze the case where y_t is predicted by x_t ? 1 with x_t being first-order autoregressive, AR(1). Under some conditions, the OLS-estimated predictive coefficient is known to be biased. We analyze a predictive model where y_t is predicted by x_t ? 1, x_t ? 2,… x_t ? p with x_t being autoregressive of order p, AR(p) with p > 1. We develop a generalized augmented regression method that produces a reduced-bias point estimate of the predictive coefficients and derive an appropriate hypothesis testing procedure. We apply our method to the prediction of quarterly stock returns by dividend yield, which is apparently AR(2). Using our method results in the AR(2) predictor series having insignificant effect, although under OLS, or the commonly assumed AR(1) structure, the predictive model is significant. We also generalize our method to the case of multiple AR(p) predictors.     

Jackknifing stock return predictions

We show that the general bias-reducing technique of jackknifing can be successfully applied to stock return predictability regressions. Compared to standard OLS estimation, the jackknifing procedure delivers virtually unbiased estimates with mean squared errors that generally dominate those of the OLS estimates. The jackknifing method is very general, as well as simple to implement, and can be applied to models with multiple predictors and overlapping observations. Unlike most previous work on inference in predictive regressions, no specific assumptions regarding the data generating process for the predictors are required. A set of Monte Carlo experiments show that the method works well in finite samples and the empirical section finds that out-of-sample forecasts based on the jackknife estimates tend to outperform those based on the plain OLS estimates. The improved forecast ability also translates into economically relevant welfare gains for an investor who uses the predictive regression, with jackknife estimates, to time the market.     

The Stambaugh bias in panel predictive regressions

This paper analyzes predictive regressions in a panel data setting. The standard fixed effects estimator suffers from a small sample bias, which is the analogue of the Stambaugh bias in time-series predictive regressions. Monte Carlo evidence shows that the bias and resulting size distortions can be severe. A new bias-corrected estimator is proposed, which is shown to work well in finite samples and to lead to approximately normally distributed t-statistics. Overall, the results show that the econometric issues associated with predictive regressions when using time-series data to a large extent also carry over to the panel case. The results are illustrated with an application to predictability in international stock indices.     

The disparity between long-term and short-term forecasted earnings growth

We find the disparity between long-term and short-term analyst forecasted earnings growth is a robust predictor of future returns and long-term analyst forecast errors. After adjusting for industry characteristics, stocks whose long-term earnings growth forecasts are far above or far below their implied short-term forecasts for earnings growth have negative and positive subsequent risk-adjusted returns along with downward and upward revisions in long-term forecasted earnings growth, respectively. Additional results indicate that investor inattention toward firm-level changes in long-term earnings growth is responsible for these risk-adjusted returns.     

Fully modified estimation with nearly integrated regressors

I show that the test procedure derived by Campbell and Yogo [2006. Journal of Financial Economics 81, 27–60] for regressions with nearly integrated variables can be interpreted as the natural t-test resulting from a fully modified estimation with near-unit-root regressors. This clearly establishes the methods of Campbell and Yogo as an extension of previous unit-root results and shows that the optimality of their test is valid under more general assumptions than those given in the original paper.     

Investor sentiment and return predictability of disagreement

This study examines the influence of investor sentiment on the relationship between disagreement among investors and future stock market returns. We find that the relationship between disagreement and future stock market returns time-varies with the degree of investor sentiment. Higher disagreement among investors’ opinions predicts significantly lower future stock market returns during high-sentiment periods, but it has no significant effect on future stock market returns during low-sentiment periods. Our findings imply that investor sentiment is related to several causes of short-sale impediments suggested in the previous literature on investor sentiment, and that the stock return predictability of disagreement is driven by investor sentiment. We demonstrate that investor sentiment has a significant impact on the stock market return predictability of disagreement through in-sample and out-of-sample analyses. In addition, the profitability of our suggested trading strategy exploiting disagreement and investor sentiment level confirms the economic significance of incorporating investor sentiment into the relationship between disagreement among investors and future stock market returns.     

Dividend growth and return predictability: A long-run re-examination of conventional wisdom

We re-examine dividend growth and return predictability evidence using 165 years of data from the Brussels Stock Exchange. The conventional wisdom holds that time-varying dividend yield is predominately explained by changes in expected returns and that expected dividend growth is only weakly forecastable. However, we find robust dividend growth predictability evidence in every time period. A lack of dividend smoothing is the most important reason for the disconnect with previous evidence. Furthermore, we find return predictability in the post-World War II period when we adjust the dividend yields for changing index composition, business cycle variation and structural breaks. This is explained by a simultaneous increase in equity duration, induced by an increasing importance of growth stocks.     

Hysteresis bands on returns,holding period and transaction costs

In the presence of transactions costs, no matter how small, arbitrage activity does not necessarily render equal all riskless rates of return. When two such rates follow stochastic processes, it is not optimal immediately to arbitrage out any discrepancy that arises between them. The reason is that immediate arbitrage would induce a definite expenditure of transactions costs whereas, without arbitrage intervention, there exists some, perhaps sufficient, probability that these two interest rates will come back together without any costs having been incurred. Hence, one can surmise that at equilibrium the financial market will permit the coexistence of two riskless rates that are not equal to each other. For analogous reasons, randomly fluctuating expected rates of return on risky assets will be allowed to differ even after correction for risk, leading to important violations of the Capital Asset Pricing Model. The combination of randomness in expected rates of return and proportional transactions costs is a serious blow to existing frictionless pricing models.     

Contrarian flows,consumption and expected stock returns

We investigate the relation between contrarian flows, consumption growth, and market risk premium. We construct a contrarian flows measure by summing up the capital flows to stocks that go against the total flow of the aggregate market. We show that the contrarian flows are negatively influenced by the same-quarter consumption growth. During bad times, the majority of investors who are affected by the negative shock reduce their equity exposure, and these extra supplies of risky assets are absorbed by contrarian investors who are least affected by the consumption shock. Using quarterly stock market data, we find that the contrarian flows forecast market returns at short-to-intermediate horizons. The predictability stems from the component that is explained by the consumption growth, and therefore the consumption growth contains valuable information about the market risk premium. Moreover, the predictability is stronger for growth stocks than for value stocks, and hence it negatively predicts the value premium. This is because the contrarian flows measure the market risk premium and growth stocks bear more discount rate risk than value stocks. Out-of-sample tests show that the main results are robust to data-snooping bias.     

As told by the supplier: Trade credit and the cross section of stock returns

With superior information about their customers’ prospects, suppliers extend trade credit to capture future profitable business. We show that this information advantage generates significant return predictability. After controlling for major firm characteristics, firms that rely more on trade credit relative to debt financing have higher subsequent stock returns. The return predictability by trade credit is stronger among firms with lower borrowing capacity or profitability, and is more significant for firms with a higher degree of information asymmetry. Our findings suggest that trade credit extension reveals suppliers’ information that diffuses gradually across the investing public.     

The option to stock volume ratio and future returns

We examine the information content of option and equity volumes when trade direction is unobserved. In a multimarket asymmetric information model, equity short-sale costs result in a negative relation between relative option volume and future firm value. In our empirical tests, firms in the lowest decile of the option to stock volume ratio (O/S) outperform the highest decile by 0.34% per week (19.3% annualized). Our model and empirics both indicate that O/S is a stronger signal when short-sale costs are high or option leverage is low. O/S also predicts future firm-specific earnings news, consistent with O/S reflecting private information.     

The term structures of equity and interest rates

This paper proposes a dynamic risk-based model capable of jointly explaining the term structure of interest rates, returns on the aggregate market, and the risk and return characteristics of value and growth stocks. Both the term structure of interest rates and returns on value and growth stocks convey information about how the representative investor values cash flows of different maturities. We model how the representative investor perceives risks of these cash flows by specifying a parsimonious stochastic discount factor for the economy. Shocks to dividend growth, the real interest rate, and expected inflation are priced, but shocks to the price of risk are not. Given reasonable assumptions for dividends and inflation, we show that the model can simultaneously account for the behavior of aggregate stock returns, an upward-sloping yield curve, the failure of the expectations hypothesis, and the poor performance of the capital asset pricing model.     

Trend following with momentum versus moving averages: a tale of differences

Despite the ever-growing interest in trend following and a series of publications in academic journals, there is a dearth of theoretical results on the properties of trend-following rules. Our paper fills this gap by comparing and contrasting the two most popular trend-following rules, the momentum (MOM) and moving average (MA) rules, from a theoretical perspective. We provide theoretical results on the similarity between different trend-following rules and the forecast accuracy of trading rules. Our results show that the similarity between the MOM and MA rules is high and increases with the strength of the trend. However, compared to the MOM rule, the MA rules exhibit more robust forecast accuracy for the future direction of price trends. In this paper, we also develop a hypothesis about uncertain market dynamics. We show that this hypothesis, coupled with our analytical results, has far-reaching practical implications and can explain a number of empirical observations. Among other things, our hypothesis explains why the empirical performance of the MA rules is better than that of the MOM rule. We broaden the appeal and practical importance of our theoretical results by offering various illustrations and real-world examples.     

Good volatility,bad volatility,and the cross section of cryptocurrency returns

This paper examines the predictability of realized volatility measures (RVM), especially the realized signed jumps (RSJ), on future volatility and returns. We confirm the existence of volatility persistence and future volatility is more strongly related to the volatility of past positive returns than to that of negative returns in the cryptocurrency market. RSJ-sorted cryptocurrency portfolios yield statistically and economically significant differences in the subsequent portfolio returns. After controlling for cryptocurrency market characteristics and existing risk factors, the differences remain significant. The investor attention explains the predictability of realized jump risk in future cryptocurrency returns.     

European business cycles and stock return predictability

This paper finds that the European leading economic indicator, a prime business cycle indicator for the European economies published by the OECD, can strongly predict European stock returns and generate utility gains. Importantly, the predictive power of the European indicator is above and beyond that contained in the country-specific leading indicator. Furthermore, we find that the predictive power of the European indicator is stable.     

Time series momentum and moving average trading rules

We compare and contrast time series momentum (TSMOM) and moving average (MA) trading rules so as to better understand the sources of their profitability. These rules are closely related; however, there are important differences. TSMOM signals occur at points that coincide with a MA direction change, whereas MA buy (sell) signals only require price to move above (below) a MA. Our empirical results show MA rules frequently give earlier signals leading to meaningful return gains. Both rules perform best outside of large stock series which may explain the puzzle of their popularity with investors, yet lack of supportive evidence in academic studies.     

Predicting firm stock returns with customer stock returns: Moderating effects of customer characteristics

Based on the multiple regression model, this study examines the potential predictive effect of customer stock returns to firm stock returns and the moderating effect of diverse customer characteristics on the predictability. By using a sample of Chinese A-share manufacturing firms listed on the Shanghai stock exchange and Shenzhen stock exchange between 2009 and 2017, we find that customer stock returns positively predict firm stock returns in the subsequent month. Additional examinations reveal that the positive predictive effect of customer stock returns on firm stock returns is more intense for firm with high proportion of state-owned customers, customer stability, customer bargaining power and customer concentration than for those with low indicators. Overall, this study contributes to the growing literature on supply chain and predictability of stock returns by shedding light on the forecasting effect of customer stock returns on firm stock returns and the predictive heterogeneity owing to customer characteristics.     

The dual contributions of information instruments in return models: magnitude and direction predictability

By partitioning asset return prediction errors, we show explicitly the dual role of magnitude and sign prediction of return instruments. We demonstrate analytically that sign prediction directly affects heteroskedasticity in asset returns; increases in precision attenuate the heteroskedasticity. Our findings with monthly asset returns are consistent with earlier evidence and indicate that our proposed analytical model captures the sign predictive component of returns. Our results are supportive of a nonlinear return generating model that can be thought of as the product of a model, perhaps linear, for forecasting return signs and a model for forecasting return magnitudes.     

Investor attention, overconfidence and category learning

Motivated by psychological evidence that attention is a scarce cognitive resource, we model investors’ attention allocation in learning and study the effects of this on asset-price dynamics. We show that limited investor attention leads to category-learning behavior, i.e., investors tend to process more market and sector-wide information than firm-specific information. This endogenous structure of information, when combined with investor overconfidence, generates important features observed in return comovement that are otherwise difficult to explain with standard rational expectations models. Our model also demonstrates new cross-sectional implications for return predictability.