A STATE‐SPACE MODEL OF SHORT‐ AND LONG‐HORIZON STOCK RETURNS

In this article we propose a new parsimonious state‐space model in which state variables characterize the stochastic movements of stock returns. Using the equally weighted and decile monthly stock returns, we show that (a) a parsimonious state‐space model characterizes the variation in expected returns at any horizon; (b) the extracted expected returns explain a substantial proportion of the variance in realized returns, and the magnitude of this proportion increases significantly with the horizon of returns; (c) the model successfully captures the empirical fact that returns of smaller firms have both stronger positive autocorrelations of short‐horizon returns and stronger negative autocorrelations of long‐horizon returns; and (d) the forecasts of asset returns obtained with the state‐space model subsume the information in other potential predictor variables such as dividend yields. JEL classification: G10, G12.     

Prediction of Individual Bond Prices via a Dynamic Bond Pricing Model: Application to Japanese Government Bond Price Data

In this paper, we propose a dynamic bond pricing model and report the usefulness of our bond pricing model based on analysis of Japanese Government bond price data. We extend the concept of the time dependent Markov (TDM) model proposed by Kariya and Tsuda (Financial Engineering and the Japanese Markets, Kluwer Academic Publishers, Dordrecht, The Netherlands, Vol. 1, pp. 1–20) to a dynamic model, which can obtain information for future bond prices. A main feature of the extended model is that the whole stochastic process of the random cash-flow discount functions of each individual bond has a time series structure. We express the dynamic structure for the models by using a Bayesian state space representation. The state space approach integrates cross-sectional and time series aspects of individual bond prices. From the empirical results, we find useful evidence that our model performs well for the prediction of the patterns of the term structure of the individual bond returns.     

Canonical Term‐Structure Models with Observable Factors and the Dynamics of Bond Risk Premia

We derive a canonical representation for the no‐arbitrage discrete‐time term structure models with both observable and unobservable state variables, popularized by Ang and Piazzesi (2003) . We conduct a specification analysis based on this canonical representation and we analyze how alternative parameterizations affect estimated risk premia, impulse response functions, and variance decompositions. We find a trade‐off between the need to obtain parsimonious parameterizations and the ability of the models to match observed patterns of variation in risk premia. We also find that more richly parameterized models uncover a greater influence of macroeconomic fundamentals on the long‐end of the yield curve.     

Applying Wand and Weber's Surface and Deep Structure Approaches to Financial Reporting Systems

Wand and Weber's fundamental premise is that ‘a physical‐symbol system has the necessary and sufficient properties to represent real‐world meaning’. Their representation of real‐world meaning flows from three possible information system models: representational, state‐tracking and de‐compositional. We address these three types of information system in the context of financial reporting systems and we use purchased goodwill and other intangible assets to selectively critique their characteristics. The principle feature of this critique is the comparison that is made throughout the paper between an economic and an artefactual (physical‐symbol) representation of assets, particularly the intangible ones, in the financial reporting domain.     

European Momentum Strategies,Information Diffusion,and Investor Conservatism

In this paper we conduct an out‐of‐sample test of two behavioural theories that have been proposed to explain momentum in stock returns. We test the gradual‐information‐diffusion model of Hong and Stein (1999) and the investor conservatism bias model of Barberis et al. (1998) in a sample of 13 European stock markets during the period 1988 to 2001. These two models predict that momentum comes from the (i) gradual dissemination of firm‐specific information and (ii) investors’ failure to update their beliefs sufficiently when they observe new public information. The findings of this study are consistent with the predictions of the behavioural models of Hong and Stein's (1999) and Barberis et al. (1998) . The evidence shows that momentum is the result of the gradual diffusion of private information and investors’ psychological conservatism reflected on the systematic errors they make in forming earnings expectations by not updating them adequately relative to their prior beliefs and by undervaluing the statistical weight of new information.     

MODELING THE TERM STRUCTURE FROM THE ON‐THE‐RUN TREASURY YIELD CURVE

We propose a new model to estimate the term structure of interest rates using observed on‐the‐run Treasury yields. The new model is an improvement over models that require a priori knowledge of the shape of the yield curve to estimate the term structure. The general form of the model is an exponential function that depends on the estimation of four parameters fit by nonlinear least squares and has straightforward interpretations. In comparing the proposed model with current yield‐curve‐smoothing models, we find that, for the data used, the proposed model does best overall in terms of pricing accuracy both in sample and out of sample. JEL classification: E43, G12     

On Testing the Random-Walk Hypothesis: A Model-Comparison Approach

The main intention of this paper is to investigate, with new daily data, whether prices in the two Chinese stock exchanges (Shanghai and Shenzhen) follow a random‐walk process as required by market efficiency. We use two different approaches, the standard variance‐ratio test of Lo and MacKinlay (1988) and a model‐comparison test that compares the ex post forecasts from a NAÏVE model with those obtained from several alternative models: ARIMA, GARCH and the Artificial Neural Network (ANN). To evaluate ex post forecasts, we utilize several procedures including RMSE, MAE, Theil's U, and encompassing tests. In contrast to the variance‐ratio test, results from the model‐comparison approach are quite decisive in rejecting the random‐walk hypothesis in both Chinese stock markets. Moreover, our results provide strong support for the ANN as a potentially useful device for predicting stock prices in emerging markets.     

Aggregation of exponential smoothing processes with an application to portfolio risk evaluation

In this paper we propose a unified framework to analyse contemporaneous and temporal aggregation of a widely employed class of integrated moving average (IMA) models. We obtain a closed-form representation for the parameters of the contemporaneously and temporally aggregated process as a function of the parameters of the original one. These results are useful due to the close analogy between the integrated GARCH (1, 1) model for conditional volatility and the IMA (1, 1) model for squared returns, which share the same autocorrelation function. In this framework, we present an application dealing with Value-at-Risk (VaR) prediction at different sampling frequencies for an equally weighted portfolio composed of multiple indices. We apply the aggregation results by inferring the aggregate parameter in the portfolio volatility equation from the estimated vector IMA (1, 1) model of squared returns. Empirical results show that VaR predictions delivered using this suggested approach are at least as accurate as those obtained by applying standard univariate methodologies, such as RiskMetrics.     

A New Econometric Model of Index Arbitrage

This paper introduces a new econometric model of the mispricing associated with (contemporaneous) differences between spot and futures prices. Like existing models, this model assumes that the level of arbitrage activity is positively related to the magnitude of absolute mispricing. However, unlike existing models, the new model assumes that a parameter governing a key feature of this relationship varies over time. Specifically, several versions of a smooth transition model of mispricing are introduced that each allow the shape of the transition function to be determined by a set of explanatory variables. Using high frequency data from the S&P 500 spot and futures market, the results show that the nature of the non‐linearity in mispricing corresponds to arbitrageur behaviour that varies (in a periodic fashion) over the trading day. This is evinced by the superior fit of the new model of mispricing, in comparison to the results based on existing econometric models of mispricing. Finally, the observed periodicity in arbitrageur behaviour indicates that arbitrageurs prefer to trade during certain periods within the trading day – a result that contradicts the findings obtained when using existing econometric models of mispricing.     

Predicting SME loan delinquencies during recession using accounting data and SME characteristics: The case of Greece

The objective of this paper is the comparison of various credit‐scoring models (i.e. binomial logistic regression, decision tree, multilayer perceptron neural network, radial basis function, and support vector machine) in evaluating the risk of small and micro enterprises' (SMEs') loan delinquencies based on accounting data and applicants' specific attributes. Exploiting a representative large data set of SMEs' loans granted by a large Greek commercial bank in the expansion period, we track the evolution of SMEs' delinquencies over the recession period August 2010 to July 2012. This time frame encompasses a period of manageable levels of delays (early recession period: August 2011–July 2012) and a period when delays were increased to a very high degree (deep recession period: August 2011–July 2012). Comparison of the employed credit‐scoring models during the early recession period shows that the multilayer perceptron neural network produces the highest predicting capacity, followed by the support vector machine model. As the crisis deepens, the support vector machine model presents the highest predicting accuracy, followed by the decision tree and then the multilayer perceptron model. Generally, the predictive performance of all credit‐scoring models seems to be substantially reduced as the recession escalates. Our paper has important implications for the proper financing of SMEs given their importance for the European economy.     

Valuation in US Commercial Real Estate

We consider a log‐linearized version of a discounted rents model to price commercial real estate as an alternative to traditional hedonic models. First, we verify a key implication of the model, namely, that cap rates forecast commercial real estate returns. We do this using two different methodologies: time series regressions of 21 US metropolitan areas and mixed data sampling (MIDAS) regressions with aggregate REIT returns. Both approaches confirm that the cap rate is related to fluctuations in future returns. We also investigate the provenance of the predictability. Based on the model, we decompose fluctuations in the cap rate into three parts: (i) local state variables (demographic and local economic variables); (ii) growth in rents; and (iii) an orthogonal part. About 30% of the fluctuation in the cap rate is explained by the local state variables and the growth in rents. We use the cap rate decomposition into our predictive regression and find a positive relation between fluctuations in economic conditions and future returns. However, a larger and significant part of the cap rate predictability is due to the orthogonal part, which is unrelated to fundamentals. This implies that economic conditions, which are also used in hedonic pricing of real estate, cannot fully account for future movements in returns. We conclude that commercial real estate prices are better modelled as financial assets and that the discounted rent model might be more suitable than traditional hedonic models, at least at an aggregate level.     

A SIMULATION ANALYSIS OF HERDING AND UNIFRACTAL SCALING BEHAVIOUR

We model the financial market using a class of agent‐based models in which agents’ expectations are driven by heuristic forecasting rules (in contrast to the rational expectations models used in traditional theories of financial markets). We show that, within this framework, we can reproduce unifractal scaling with respect to three well‐known power laws relating (i) moments of the absolute price change to the time‐scale over which they are measured, (ii) magnitude of returns with respect to their probability and (iii) the autocorrelation of absolute returns with respect to lag. In contrast to previous studies, we systematically analyse all three power laws simultaneously using the same underlying model by making observations at different time‐scales and higher moments. We show that the first two scaling laws are remarkably robust to the time‐scale over which observations are made, irrespective of the model configuration. However, in contrast to previous studies, we show that herding may explain why long memory is observed at all frequencies. Copyright © 2013 John Wiley & Sons, Ltd.     

The Stochastic Seasonal Behaviour of Natural Gas Prices

Previous studies have explored the seasonal behaviour of commodity prices as a deterministic factor. This paper goes further by proposing a general (n+2m)‐factor model for the stochastic behaviour of commodity prices, which nests the deterministic seasonal model by Sorensen (2002) . We consider seasonality as a stochastic factor, with n non‐seasonal and m seasonal factors. The non‐seasonal factors are as defined in Schwartz (1997) , Schwartz and Smith (2000) and Cortazar and Schwartz (2003) . The seasonal factors are trigonometric components generated by stochastic processes. The model has been applied to the Henry Hub natural gas futures contracts listed by NYMEX. We find that models allowing for stochastic seasonality outperform standard models with deterministic seasonality. We obtain similar results with other energy commodities. Moreover, we find that stochastic seasonality implies that the volatility of futures returns follows a seasonal pattern. This result has important implications in terms of option pricing.     

Risk-Neutral Parameter Shifts and Derivatives Pricing in Discrete Time

We obtain a large class of discrete‐time risk‐neutral valuation relationships, or “preference‐free” derivatives pricing models, by imposing a simple restriction on the state‐price density process. The risk‐neutral stock‐return and forward‐rate dynamics are obtained by changing only a location parameter, which can be determined independent of the preference and true location parameters. The Gaussian models of Rubinstein (1976) , Brennan (1979) , and Câmera (2003) , and the gamma model of Heston (1993) are all special cases. The model provides simple relationships between expected returns and state‐price density parameters analogous to the diffusion case.     

Bad beta good beta,state‐space news decomposition and the cross‐section of stock returns

This study employs an innovative market‐based approach, where return on equity (ROE) is employed as a proxy for cash‐flow news and a state‐space model is used for market news decomposition. We document that the bad beta good beta (BBGB) model of Campbell and Vuolteenaho (2004) explains about 30 per cent of the cross‐sectional variations in US stock returns. We also find that the BBGB model adequately explains the size effect leading to its superior performance in this area. Our method controls for the news decomposition method and market news proxies’ bias. We contribute to the literature by providing an alternative easy‐to‐implement and consistent market‐based method for news decomposition.     

The Performance of Hedge Fund Strategies and the Asymmetry of Return Distributions

We present hedge fund performance estimates that adjust for stale prices, Fama‐French risk factors and skewness. We contrast these new performance estimates with traditional performance measures. Using three‐factor models to adjust for staleness in prices and to incorporate Fama‐French factors along with the Harvey‐Siddique (2000) two‐factor model that incorporates skewness, we find that for the period 1990–2003, all hedge fund categories achieve above average performance when measured against an aggregate market index. More significantly, however, when we estimate performance at the individual hedge fund level, we discover that only 40 to 47% of the funds are shown to achieve an above average performance over that time period depending on the model used. These results have important implications for investors, endowments and pensions when they choose hedge fund managers.     

Cross‐sectional Tests of Conditional Asset Pricing Models: Evidence from the German Stock Market

We study the performance of conditional asset pricing models and multifactor models in explaining the German cross‐section of stock returns. We focus on several variables, which (according to previous research) are associated with market expectations on future market excess returns or business cycle conditions. Our results suggest that the empirical performance of the Capital Asset Pricing Model (CAPM) can be improved when allowing for time‐varying parameters of the stochastic discount factor. A conditional CAPM using the term spread explains the returns on our size and book‐to‐market sorted portfolios about as well as the Fama‐French three‐factor model and performs best in terms of the Hansen‐Jagannathan distance. Structural break tests do not necessarily indicate parameter instability of conditional model specifications. Another major finding of the paper is that the Fama‐French model – despite its generally good cross‐sectional performance – is subject to model instability. Unconditional models, however, do a better job than conditional ones at capturing time‐series predictability of the test portfolio returns.     

Estimating GARCH models using support vector machines*

Abstract

Support vector machines (SVMs) are a new nonparametric tool for regression estimation. We will use this tool to estimate the parameters of a GARCH model for predicting the conditional volatility of stock market returns. GARCH models are usually estimated using maximum likelihood (ML) procedures, assuming that the data are normally distributed. In this paper, we will show that GARCH models can be estimated using SVMs and that such estimates have a higher predicting ability than those obtained via common ML methods.     

Prediction of Operating Cash Flows: Further Evidence from Australia

This paper examines empirically the relative abilities of current operating cash flows (hereafter OCF) and earnings in predicting future operating cash flows in Australia. It extends prior Australian research on cash flow prediction ( Percy and Stokes 1992 ; Clinch, Sidhu and Sing 2002 ; Farshadfar, Ng and Brimble 2009 ) by examining future cash flow predictions for one‐, two‐ and three‐year‐ahead forecast horizons; incorporating additional contextual variables likely to affect the predictive association between current cash flows or earnings and future cash flows; and comparing cross‐sectional versus time series‐based prediction models to ascertain the relative superiority of one approach over the other. Regression results reveal that the cash flow‐based models are more accurate in predicting future operating cash flows than earnings‐based models. This result, however, is moderated by firm‐specific contextual factors like firm size, negative versus positive cash flow pattern, cash flow variability and firm operating cycle. Finally, a comparison between cross‐sectional and time series approaches reveals that the cross‐sectional model outperforms the time series model for both the operating cash flows and earnings models in most of the forecast years.     

The Sophisticated and the Simple: the Profitability of Contrarian Strategies

A variety of variables have been used to form contrarian portfolios, ranging from relatively simple measures, like book‐to‐market, cash flow‐to‐price, earnings‐to‐price and past returns, to more sophisticated measures based on the Ohlson model and residual income model (RIM). This paper investigates whether: (i) contrarian strategies based on RIM perform better or worse than those based on the Ohlson model; (ii) contrarian strategies based on more sophisticated valuation models (e.g. Ohlson and RIM) perform much better than the relatively simpler ranking variables that have been used so extensively in the finance literature. Given that the RIM and Ohlson models require greater information inputs and technical know‐how, and make different implicit assumptions on future abnormal earnings, it is important to ascertain if they offer significantly greater contrarian profits to outweigh the increased costs that they entail. Indeed, our surprising finding is that simple cash flow‐to‐price measures appear to do almost as well as the more sophisticated alternatives. One would have expected the sophisticated models to significantly outperform the simple cash flow to price model for the reasons given by Penman (2007) .