A measure of time series' predictability using genetic programming applied to stock returns

Based on the standard genetic programming (GP) paradigm, we introduce a new probability measure of time series' predictability. It is computed as a ratio of two fitness values (SSE) from GP runs. One value belongs to a subject series, while the other belongs to the same series after it is randomly shuffled. Theoretically, the boundaries of the measure are between zero and 100, where zero characterizes stochastic processes while 100 typifies predictable ones. To evaluate its performance, we first apply it to experimental data. It is then applied to eight Dow Jones stock returns. This measure may reduce model search space and produce more reliable forecast models. Copyright © 1999 John Wiley & Sons, Ltd.     

Predicting the Distribution of Stock Returns: Model Formulation,Statistical Evaluation,VaR Analysis and Economic Significance

A large literature has investigated predictability of the conditional mean of low‐frequency stock returns by macroeconomic and financial variables; however, little is known about predictability of the conditional distribution. We look at one‐step‐ahead out‐of‐sample predictability of the conditional distribution of monthly US stock returns in relation to the macroeconomic and financial environment. Our methodological approach is innovative: we consider several specifications for the conditional density and combinations schemes. Our results are as follows: the entire density is predicted under combination schemes as applied to univariate GARCH models with Gaussian innovations; the Bayesian winner in relation to GARCH‐skewed‐t models is informative about the 5% value at risk; the average realised utility of a mean–variance investor is maximised under the Bayesian winner as applied to GARCH models with symmetric Student t innovations. Our results have two implications: the best prediction model depends on the evaluation criterion; and combination schemes outperform individual models. Copyright © 2015 John Wiley & Sons, Ltd.     

Does a lot help a lot? Forecasting stock returns with pooling strategies in a data‐rich environment

A variety of recent studies provide a skeptical view on the predictability of stock returns. Empirical evidence shows that most prediction models suffer from a loss of information, model uncertainty, and structural instability by relying on low‐dimensional information sets. In this study, we evaluate the predictive ability of various lately refined forecasting strategies, which handle these issues by incorporating information from many potential predictor variables simultaneously. We investigate whether forecasting strategies that (i) combine information and (ii) combine individual forecasts are useful to predict US stock returns, that is, the market excess return, size, value, and the momentum premium. Our results show that methods combining information have remarkable in‐sample predictive ability. However, the out‐of‐sample performance suffers from highly volatile forecast errors. Forecast combinations face a better bias–efficiency trade‐off, yielding a consistently superior forecast performance for the market excess return and the size premium even after the 1970s.     

Time series multistep‐ahead predictability estimation and ranking

A predictability index was defined as the ratio of the variance of the optimal prediction to the variance of the original time series by Granger and Anderson (1976) and Bhansali (1989). A new simplified algorithm for estimating the predictability index is introduced and the new estimator is shown to be a simple and effective tool in applications of predictability ranking and as an aid in the preliminary analysis of time series. The relationship between the predictability index and the position of the poles and lag p of a time series which can be modelled as an AR(p) model are also investigated. The effectiveness of the algorithm is demonstrated using numerical examples including an application to stock prices. Copyright © 1999 John Wiley & Sons, Ltd.     

Model uncertainty,thick modelling and the predictability of stock returns

Recent financial research has provided evidence on the predictability of asset returns. In this paper we consider the results contained in Pesaran and Timmerman (1995), which provided evidence on predictability of excess returns in the US stock market over the sample 1959–1992. We show that the extension of the sample to the nineties weakens considerably the statistical and economic significance of the predictability of stock returns based on earlier data. We propose an extension of their framework, based on the explicit consideration of model uncertainty under rich parameterizations for the predictive models. We propose a novel methodology to deal with model uncertainty based on ‘thick’ modelling, i.e. on considering a multiplicity of predictive models rather than a single predictive model. We show that portfolio allocations based on a thick modelling strategy systematically outperform thin modelling. Copyright © 2005 John Wiley & Sons, Ltd.     

Forecasting Performance of Nonlinear Models for Intraday Stock Returns

We studied the predictability of intraday stock market returns using both linear and nonlinear time series models. For the S&P 500 index we compared simple autoregressive and random walk linear models with a range of nonlinear models, including smooth transition, Markov switching, artificial neural network, nonparametric kernel regression and support vector machine models for horizons of 5, 10, 20, 30 and 60 minutes. The empirical results indicate that nonlinear models outperformed linear models on the basis of both statistical and economic criteria. Specifically, although return serial correlation receded by around 10 minutes, return predictability still persisted for up to 60 minutes according to nonlinear models, even though profitability decreases as time elapses. More flexible nonlinear models such as support vector machines and artificial neural network did not clearly outperform other nonlinear models. Copyright © 2011 John Wiley & Sons, Ltd.     

Term Structure Forecasting of Government Bond Yields with Latent and Macroeconomic Factors: Do Macroeconomic Factors Imply Better Out‐of‐Sample Forecasts?

This study empirically examines the role of macroeconomic and stock market variables in the dynamic Nelson–Siegel framework with the purpose of fitting and forecasting the term structure of interest rate on the Japanese government bond market. The Nelson–Siegel type models in state‐space framework considerably outperform the benchmark simple time series forecast models such as an AR(1) and a random walk. The yields‐macro model incorporating macroeconomic factors leads to a better in‐sample fit of the term structure than the yields‐only model. The out‐of‐sample predictability of the former for short‐horizon forecasts is superior to the latter for all maturities examined in this study, and for longer horizons the former is still compatible to the latter. Inclusion of macroeconomic factors can dramatically reduce the autocorrelation of forecast errors, which has been a common phenomenon of statistical analysis in previous term structure models. Copyright © 2013 John Wiley & Sons, Ltd.     

Predicting Bid–Ask Spreads Using Long‐Memory Autoregressive Conditional Poisson Models

We introduce a long‐memory autoregressive conditional Poisson (LMACP) model to model highly persistent time series of counts. The model is applied to forecast quoted bid–ask spreads, a key parameter in stock trading operations. It is shown that the LMACP nicely captures salient features of bid–ask spreads like the strong autocorrelation and discreteness of observations. We discuss theoretical properties of LMACP models and evaluate rolling‐window forecasts of quoted bid–ask spreads for stocks traded at NYSE and NASDAQ. We show that Poisson time series models significantly outperform forecasts from AR, ARMA, ARFIMA, ACD and FIACD models. The economic significance of our results is supported by the evaluation of a trade schedule. Scheduling trades according to spread forecasts we realize cost savings of up to 14 % of spread transaction costs. Copyright © 2013 John Wiley & Sons, Ltd.     

What can we learn from the return predictability over the business cycle?

In this paper, we forecast stock returns using time‐varying parameter (TVP) models with parameters driven by economic conditions. An in‐sample specification test shows significant variation in the parameters. Out‐of‐sample results suggest that the TVP models outperform their constant coefficient counterparts. We also find significant return predictability from both statistical and economic perspectives with the application of TVP models. The out‐of‐sample R² of an equal‐weighted combination of TVP models is as high as 2.672%, and the gains in the certainty equivalent return are 214.7 basis points. Further analysis indicates that the improvement in predictability comes from the use of information on economic conditions rather than simply from allowing the coefficients to vary with time.     

Improving Hull and White's Method of Estimating Portfolio Value‐at‐Risk

We propose a method approach. We use six international stock price indices and three hypothetical portfolios formed by these indices. The sample was observed daily from 1 January 1996 to 31 December 2006. Confirmed by the failure rates and backtesting developed by Kupiec (Technique for verifying the accuracy of risk measurement models. Journal of Derivatives 1995; 3 : 73–84) and Christoffersen (Evaluating interval forecasts. International Economic Review 1998; 39 : 841–862), the empirical results show that our method can considerably improve the estimation accuracy of value‐at‐risk. Thus the study establishes an effective alternative model for risk prediction and hence also provides a reliable tool for the management of portfolios. Copyright © 2011 John Wiley & Sons, Ltd.     

Performance of GARCH models in forecasting stock market volatility

This paper studies the performance of GARCH model and its modifications, using the rate of returns from the daily stock market indices of the Kuala Lumpur Stock Exchange (KLSE) including Composite Index, Tins Index, Plantations Index, Properties Index, and Finance Index. The models are stationary GARCH, unconstrained GARCH, non‐negative GARCH, GARCH‐M, exponential GARCH and integrated GARCH. The parameters of these models and variance processes are estimated jointly using the maximum likelihood method. The performance of the within‐sample estimation is diagnosed using several goodness‐of‐fit statistics. We observed that, among the models, even though exponential GARCH is not the best model in the goodness‐of‐fit statistics, it performs best in describing the often‐observed skewness in stock market indices and in out‐of‐sample (one‐step‐ahead) forecasting. The integrated GARCH, on the other hand, is the poorest model in both respects. Copyright © 1999 John Wiley & Sons, Ltd.     

Volatility spillover from the US to international stock markets: A heterogeneous volatility spillover GARCH model

A recent study by Rapach, Strauss, and Zhou (Journal of Finance, 2013, 68(4), 1633–1662) shows that US stock returns can provide predictive content for international stock returns. We extend their work from a volatility perspective. We propose a model, namely a heterogeneous volatility spillover–generalized autoregressive conditional heteroskedasticity model, to investigate volatility spillover. The model specification is parsimonious and can be used to analyze the time variation property of the spillover effect. Our in‐sample evidence shows the existence of strong volatility spillover from the US to five major stock markets and indicates that the spillover was stronger during business cycle recessions in the USA. Out‐of‐sample results show that accounting for spillover information from the USA can significantly improve the forecasting accuracy of international stock price volatility.     

Stochastic Multivariate Mixture Covariance Model

Stochastic covariance models have been explored in recent research to model the interdependence of assets in financial time series. The approach uses a single stochastic model to capture such interdependence. However, it may be inappropriate to assume a single coherence structure at all time t. In this paper, we propose the use of a mixture of stochastic covariance models to generalize the approach and offer greater flexibility in real data applications. Parameter estimation is performed by Bayesian analysis with Markov chain Monte Carlo sampling schemes. We conduct a simulation study on three different model setups and evaluate the performance of estimation and model selection. We also apply our modeling methods to high‐frequency stock data from Hong Kong. Model selection favors a mixture rather than non‐mixture model. In a real data study, we demonstrate that the mixture model is able to identify structural changes in market risk, as evidenced by a drastic change in mixture proportions over time. Copyright © 2016 John Wiley & Sons, Ltd.     

Forecasting the recent behavior of US business fixed investment spending: an analysis of competing models

We evaluate forecasting models of US business fixed investment spending growth over the recent 1995:1–2004:2 out‐of‐sample period. The forecasting models are based on the conventional Accelerator, Neoclassical, Average Q, and Cash‐Flow models of investment spending, as well as real stock prices and excess stock return predictors. The real stock price model typically generates the most accurate forecasts, and forecast‐encompassing tests indicate that this model contains most of the information useful for forecasting investment spending growth relative to the other models at longer horizons. In a robustness check, we also evaluate the forecasting performance of the models over two alternative out‐of‐sample periods: 1975:1–1984:4 and 1985:1–1994:4. A number of different models produce the most accurate forecasts over these alternative out‐of‐sample periods, indicating that while the real stock price model appears particularly useful for forecasting the recent behavior of investment spending growth, it may not continue to perform well in future periods. Copyright © 2007 John Wiley & Sons, Ltd.     

Forecasting and trading currency volatility: an application of recurrent neural regression and model combination

In this paper, we examine the use of non‐parametric Neural Network Regression (NNR) and Recurrent Neural Network (RNN) regression models for forecasting and trading currency volatility, with an application to the GBP/USD and USD/JPY exchange rates. Both the results of the NNR and RNN models are benchmarked against the simpler GARCH alternative and implied volatility. Two simple model combinations are also analysed. The intuitively appealing idea of developing a nonlinear nonparametric approach to forecast FX volatility, identify mispriced options and subsequently develop a trading strategy based upon this process is implemented for the first time on a comprehensive basis. Using daily data from December 1993 through April 1999, we develop alternative FX volatility forecasting models. These models are then tested out‐of‐sample over the period April 1999–May 2000, not only in terms of forecasting accuracy, but also in terms of trading efficiency: in order to do so, we apply a realistic volatility trading strategy using FX option straddles once mispriced options have been identified. Allowing for transaction costs, most trading strategies retained produce positive returns. RNN models appear as the best single modelling approach yet, somewhat surprisingly, model combination which has the best overall performance in terms of forecasting accuracy, fails to improve the RNN‐based volatility trading results. Another conclusion from our results is that, for the period and currencies considered, the currency option market was inefficient and/or the pricing formulae applied by market participants were inadequate. Copyright © 2002 John Wiley & Sons, Ltd.     

Out‐of‐sample volatility prediction: A new mixed‐frequency approach

This paper proposes a new mixed‐frequency approach to predict stock return volatilities out‐of‐sample. Based on the strategy of momentum of predictability (MoP), our mixed‐frequency approach has a model switching mechanism that switches between generalized autoregressive conditional heteroskedasticity (GARCH)‐class models that only use low‐frequency data and heterogeneous autoregressive models of realized volatility (HAR‐RV)‐type that only use high‐frequency data. The MoP model simply selects a forecast with relatively good past performance between the GARCH‐class and HAR‐RV‐type forecasts. The model confidence set (MCS) test shows that our MoP strategy significantly outperforms the competing models, which is robust to various settings. The MoP test shows that a relatively good recent past forecasting performance of the GARCH‐class or HAR‐RV‐type model is significantly associated with a relatively good current performance, supporting the success of the MoP model.     

Equity return predictability,its determinants,and profitable trading strategies

This paper explains cross‐market variations in the degree of return predictability using the extreme bounds analysis (EBA). The EBA addresses model uncertainty in identifying robust determinant(s) of cross‐sectional return predictability. Additionally, the paper develops two profitable trading strategies based on return predictability evidence. The result reveals that among the 13 determinants of the cross‐sectional variation of return predictability, only value of stock traded (a measure of liquidity) is found to have robust explanatory power by Leamer's (1985) EBA. However, Sala‐i‐Martin's (1997) EBA reports that value of stock traded, gross domestic product (GDP) per capita, level of information and communication technology (ICT) development, governance quality, and corruption perception are robust determinants. We further find that a strategy of buying (selling) aggregate market portfolios of the countries with the highest positive (negative) return predictability statistic in the past 24 months generates statistically significant positive returns in the subsequent 3 to 12 months. In the individual country level, a trading rule of buying (selling) the respective country's aggregate market portfolio, when the return predictability statistic turns out positive (negative), outperforms the conventional buy‐and‐hold strategy for many countries.     

Cross‐correlations and predictability of stock returns

Studies have shown that small stock returns can be partially predicted by the past returns of large stocks (cross‐correlations), while a larger body of literature has shown that macroeconomic variables can predict future stock returns. This paper assesses the marginal contribution of cross‐correlations after controlling for predictability inherent in lagged macroeconomic variables. Macroeconomic forecasting models generate trading rule profits of up to 0·431% per month, while the inclusion of cross‐correlations increases returns to 0·516% per month. Such results suggest that cross‐correlations may serve as a proxy for omitted macroeconomic variables in studies of stock market predictability. Macroeconomic variables are more important than cross‐correlations in forecasting small stock returns and encompassing tests suggest that the small marginal contribution of cross‐correlations is not statistically significant. Copyright © 2000 John Wiley & Sons, Ltd.     

Moment tests for density forecast evaluation in the presence of parameter estimation uncertainty

Density forecast (DF) possesses appealing properties when it is correctly specified for the true conditional distribution. Although a number of parametric specification tests have been introduced for the DF evaluation (DFE) in the parameter‐free context, econometric DF models are typically parameter‐dependent. In this paper, we first use a generalized probability integral transformation‐based moment test to unify these existing tests, and then apply the Newey–Tauchen method (the West–McCracken method) to correct this unified test as a generalized full‐sample (out‐of‐sample) test in the parameter‐dependent context. Unlike the corrected tests, the uncorrected tests could be substantially undersized (oversized) when they are directly applied to the full‐sample (out‐of‐sample) DFE in the presence of parameter estimation uncertainty. We also use a simulation to show the usefulness of the corrected tests in rectifying the size distortion problem, and apply the corrected tests to an empirical study of stock index returns. Copyright © 2010 John Wiley & Sons, Ltd.     

Forecasting commercial paper rates

A model previously developed by Lackman (C. L. Lackman, Forecasting commercial paper rates. Journal of Business Finance and Accounting 15 (1988) 499–524) for the period 1960 to 1985 is updated to include the 1990s and incorporate statistical techniques relating to tests for stationary conditions not available in 1988. As in the previous model, the demand for commercial paper by each institution (Households (HH), Life Insurance Companies (LIC), Non‐Financial Corporations (CRP) and Finance Corporations (FC)) and the total demand is simulated. Simulations of the commercial paper rate are also generated—using just the demand equations (total supply exogenous) and then employing the entire model (supply endogenous) to determine the rate. Simulation periods are from 1960:2 to 2001:4 for all demand simulations. The dynamic simulation of the total demand for commercial paper performs well. The resulting root mean square error, 3.485, compares favourably with the Federal Reserve Boston–Massachusetts Institute of Technology (FRB–MIT) estimate of the commercial paper rate (deLeeuw and Granlich, 1968). Copyright © 2004 John Wiley & Sons, Ltd.