Interpreting Factor Models

We argue that tests of reduced‐form factor models and horse races between “characteristics” and “covariances” cannot discriminate between alternative models of investor beliefs. Since asset returns have substantial commonality, absence of near‐arbitrage opportunities implies that the stochastic discount factor can be represented as a function of a few dominant sources of return variation. As long as some arbitrageurs are present, this conclusion applies even in an economy in which all cross‐sectional variation in expected returns is caused by sentiment. Sentiment‐investor demand results in substantial mispricing only if arbitrageurs are exposed to factor risk when taking the other side of these trades.     

Integrating Neural Networks for Risk-Adjustment Models

This article demonstrates the possibility of an alternative approach for risk-adjustment models. In the proposed model the risk characteristics of the beneficiary's health within the same cohort classified by Self-Organizing Map network are highly homogeneous, whereas the numbers of individuals within each cohort remain sufficient to allow further investigation of the causal effect from clustered data. A comparison of different models by the 10-fold cross-validation reveals that the performance improvement in the proposed integration model is both significant and stable across the estimation and validation sampling.     

Factor Models for Option Pricing

Options on stocks are priced using information on index options and viewing stocks in a factor model as indirectly holding index risk. The method is particularly suited to developing quotations on stock options when these markets are relatively illiquid and one has a liquid index options market to judge the index risk. The pricing strategy is illustrated on IBM and Sony options viewed as holding SPX and Nikkei risk respectively.     

Identification of Factor Models for Forecasting Returns

A data-driven approach for forecasting returns of asset pricesis introduced. Special emphasis is given to data-driven specificationand to dimension reduction. Specification is performed by amodified AIC, BIC-based An-algorithm. Quasi-static principalcomponent analysis, quasi-static factor models with idiosyncraticerrors and reduced rank regression are considered. The forecastingresults obtained are compared.     

Efficient Factor Models For Yield Curve Dynamics

Abstract

This paper derives a class of efficient factor models that bridge a gap between factor models and Heath-Jarrow-Morton models. These efficient factor models provide arbitrage-free dynamics for the yield curve, can be readily extended to fit the current yield curve, and have closed-form formulas for pricing default-free zero-coupon bonds. The short rate is a state variable in these efficient factor models. There are no restrictions imposed on the functional form of the volatility of the short rate except for certain technical conditions to ensure the solvability of the associated stochastic differential equations. The stochastic volatility of the short rate can be one of the state variables. The paper also presents a closed-form solution for default-free discount bond prices in the Malkiel model and provides a new method to derive the Ritchken-Sankarasubramanian model.     

The Stability of Factor Models of Interest Rates

The daily term structure of interest rates is filtered to reducethe influence of cross-correlations and autocorrelations onits factors. A three-factor model is fitted to the filtereddata. We perform statistical tests, finding that factor loadingsare unstable through time for daily data. This finding is notdue to the presence of outliers nor to the selected number offactors. Such an instability problem can be solved when applyingthe factor analysis on multivariate scaled residuals, filteredusing a nonparametric technique based on functional gradientdescent.     

Assessing Specification Errors in Stochastic Discount Factor Models

In this article we develop alternative ways to compare asset pricing models when it is understood that their implied stochastic discount factors do not price all portfolios correctly. Unlike comparisons based on ${\chi }^2$ statistics associated with null hypotheses that models are correct, our measures of model performance do not reward variability of discount factor proxies. One of our measures is designed to exploit fully the implications of arbitrage-free pricing of derivative claims. We demonstrate empirically the usefulness of our methods in assessing some alternative stochastic factor models that have been proposed in asset pricing literature.     

Constructing Asset Pricing Models With Specific Factor Loadings

We demonstrate how one can build pricing formulae in which factors other than beta may be viewed as determinants of asset returns. This is important conceptually as it demonstrates how the additional factors can compensate for a market portfolio proxy that is mis‐specified, and also shows how such a pricing model can be specified ex ante. The procedure is implemented by first selecting an ‘orthogonal’ portfolio which falls on the mean‐variance efficient frontier computed from the empirical average returns, variances and covariances on the equity securities of a large sample of firms. One then determines the inefficient index portfolio which leads to a vector of betas that when multiplied by the average return on the orthogonal portfolio, and which when subtracted from the vector of average returns for the firms comprising the sample, yields an error vector that is equal to the vector of numerical values for the variables that are to form the basis of the asset pricing formula. There will then be a perfect linear relationship between the vector of average returns for the firms comprising the sample, the vector of betas based on the inefficient index portfolio and such other factors that are deemed to be important in the asset pricing process. We illustrate computational procedures using a numerical example based on the quality of information contained in published corporate financial statements.     

Integrating Credit Models Using Accounting Information With Loan Officers' Decision Processes

This paper focuses on loan officers' cognitive processes of problem detection and problem-hypothesis formulation, which are incorporated into a credit model when they are confronted with loan decisions. Prior credit models in banking have not directly addressed loan officers' internal processes in a loan situation. The integration of both loan officers' cognitive processes and information used in a credit model can better help explain their decision-making biases. The results presented in this paper showed that information derived from a credit model influences loan officers' problem detection and problem-hypothesis formulation, and these processes are important factors in their loan approval. To identify loan officers' decision-making processes, the approach used in this paper integrates principles from financial management, economics, and cognitive psychology with methodological developments from psychometrics and econometrics.     

Dynamic Investment Strategy with Factor Models Under Regime Switches

A model for dynamic investment strategy is developed where assets’ returns are represented by multiple factors. In a mean–variance framework with factor models under regime switches, we derive a semi-analytic solution for the optimal portfolio with transaction costs. Due to the existence of transaction costs, the optimal portfolio is characterized as a linear combination of current and target portfolios, the latter of which maximizes the value function in the current regime. For some special cases of interest, we also derive simplified analytical solutions. To see the effect of regime switches, the proposed model is applied to US equity market in which small minus big and high minus low are employed as factors. Investment strategy based on our model demonstrates empirically that the regime switching models exhibit superior performance over the single regime model for such performance measures as realized utility and Sharpe ratio which are of particular interest in practice. Taking a close look at the time series of portfolio returns, the result shows the usefulness of the regime switching model as investors flexibly optimize asset allocations depending on the state of the market.     

An Examination of Alternative Factor Models in UK Stock Returns

This paper examines the mean-variance efficiency of a number offactor models in UK stock returns. The paper also explores, using theapproach of MacKinlay (1995), whether missing risk factors ornonrisk-based explanations best explain the pricing errors of thedifferent factor models. The evidence in the paper suggests that themean-variance efficiency of each factor model is rejected and missing riskfactors are unable to explain the pricing errors of any of the models.Some nonrisk-based explanations, which posit a wide spread in abnormalreturns, may be a more plausible source of explaining the pricing errorsof the factor models.     

Determining an Optimal Principal Limit Factor for Reverse Mortgages under Economics-Based Models

The Journal of Real Estate Finance and Economics - Determining an optimal principal limit factor (PLF) is important for a reverse mortgage (RM) contract because it mainly influences the development...     

Arbitrage and the Evaluation of Linear Factor Models in UK Stock Returns

I examine the impact of the no arbitrage restriction on the estimation and evaluation of linear factor models in UK stock returns. The no arbitrage restriction reduces volatility and eliminates most of the negative values of the fitted stochastic discount factor models. All of the factor models are rejected and there are significant differences in the pricing performance between models under the no arbitrage restriction. The no arbitrage restriction can have a significant impact on both the parameter estimates and pricing errors for some models.     

Bayesian Solutions for the Factor Zoo: We Just Ran Two Quadrillion Models

We propose a novel framework for analyzing linear asset pricing models: simple, robust, and applicable to high-dimensional problems. For a (potentially misspecified) stand-alone model, it provides reliable price of risk estimates for both tradable and nontradable factors, and detects those weakly identified. For competing factors and (possibly nonnested) models, the method automatically selects the best specification—if a dominant one exists—or provides a Bayesian model averaging–stochastic discount factor (BMA-SDF), if there is no clear winner. We analyze 2.25 quadrillion models generated by a large set of factors and find that the BMA-SDF outperforms existing models in- and out-of-sample.     

CAPM, Higher Co‐moment and Factor Models of UK Stock Returns

In this paper we examine the variables that explain the cross‐section of UK stock returns. Previous studies have found that the CAPM beta has moderate or even insignificant explanatory power once the Fama French factors are included. However, we control for different realised risk premia in up and down markets by using the same methodology as Pettengill, Sundaram and Mathur (1995). Unlike previous work, we find that beta is highly significant in explaining the cross‐section of UK stock returns and more importantly remains significant even when the Fama French factors are included in the cross‐sectional regressions. We also investigate whether higher co‐moments (co‐skewness and co‐kurtosis) have any explanatory power but find that empirical support is weaker.     

Forecasting the Correlation Structure of German Stock Returns: A Test of Firm-Specific Factor Models

This paper evaluates the performance of various factor models with firm-specific variables in forecasting correlation matrices at the German stock market. We investigate forecasts of correlations for a comprehensive sample and a sample of blue chips and analyse the impact of stock market crashes on the forecasting accuracy. Our empirical results show that the multi-factor models do not generally produce better forecasts than 'naive' models. Specifically, the traditional industry mean model significantly outperforms all other techniques in most of the time periods.