[下载] Linear Factor Models in Finance

4073

收藏 2008-12-19

Linear Factor Models in Finance John Knight and Stephen SatchellElsevier Finance Contents List of contributors xi Introduction xv 1 Review of literature on multifactor asset pricing models 1 Mario Pitsillis 1.1 Theoretical reasons for existence of multiple factors 1 1.2 Empirical evidence of existence of multiple factors 5 1.3 Estimation of factor pricing models 5 Bibliography 9 2 Estimating UK factor models using the multivariate skew normal distribution 12 C. J. Adcock 2.1 Introduction 12 2.2 The multivariate skew normal distribution and some of its properties 14 2.3 Conditional distributions and factor models 17 2.4 Data model choice and estimation 19 2.5 Empirical study 19 2.5.1 Basic return statistics 19 2.5.2 Overall model fit 21 2.5.3 Comparison of parameter estimates 23 2.5.4 Skewness parameters 24 2.5.5 Tau and time-varying conditional variance 25 2.6 Conclusions 27 Acknowledgement 27 References 27 3 Misspecification in the linear pricing model 30 Ka-Man Lo 3.1 Introduction 30 3.2 Framework 31 3.2.1 Arbitrage Pricing Theory 31 3.2.2 Multivariate F test used in linear factor model 32 3.2.3 Average F test used in linear factor model 34 vi Contents 3.3 Distribution of the multivariate F test statistics under misspecification 34 3.3.1 Exclusion of a set of factors from estimation 35 3.3.2 Time-varying factor loadings 41 3.4 Simulation study 43 3.4.1 Design 43 3.4.2 Factors serially independent 45 3.4.3 Factors autocorrelated 48 3.4.4 Time-varying factor loadings 49 3.4.5 Simulation results 50 3.5 Conclusion 57 Appendix: Proof of proposition 3.1 and proposition 3.2 59 4 Bayesian estimation of risk premia in an APT context 61 Theofanis Darsinos and Stephen E. Satchell 4.1 Introduction 61 4.2 The general APT framework 62 4.2.1 The excess return generating process (when factors are traded portfolios) 62 4.2.2 The excess return generating process (when factors are macroeconomic variables or non-traded portfolios) 64 4.2.3 Obtaining the (K × 1) vector of risk premia λ 65 4.3 Introducing a Bayesian framework using a Minnesota prior (Litterman’s prior) 66 4.3.1 Prior estimates of the risk premia 67 4.3.2 Posterior estimates of the risk premia 70 4.4 An empirical application 72 4.4.1 Data 73 4.4.2 Results 74 4.5 Conclusion 77 References 77 Appendix 80 5 Sharpe style analysis in the MSCI sector portfolios: a Monte Carlo integration approach 83 George A. Christodoulakis 5.1 Introduction 83 5.2 Methodology 84 5.2.1 A Bayesian decision-theoretic approach 85 5.2.2 Estimation by Monte Carlo integration 86 5.3 Style analysis in the MSCI sector portfolios 87 5.4 Conclusions 93 References 93 Contents vii 6 Implication of the method of portfolio formation on asset pricing tests 95 Ka-Man Lo 6.1 Introduction 95 6.2 Models 97 6.2.1 Asset pricing frameworks 97 6.2.2 Specifications to be tested 98 6.3 Implementation 99 6.3.1 Multivariate F test 99 6.3.2 Average F test 100 6.3.3 Stochastic discount factor using GMM with Hansen and Jagannathan distance 102 6.3.4 A look at the pricing errors under different tests 103 6.4 Variables construction and data sources 104 6.4.1 Data sources 104 6.4.2 Independent variables: excess market return, size return factor and book-to-market return factor 105 6.4.3 Dependent variables: size-sorted portfolios, beta-sorted portfolios and individual assets 109 6.5 Result and discussion 114 6.5.1 Formation of WT 114 6.5.2 Model 1 115 6.5.3 Model 2 123 6.5.4 Model 3 133 6.6 Simulation 138 6.7 Conclusion and implication 146 References 148 7 The small noise arbitrage pricing theory and its welfare implications 150 Stephen E. Satchell 7.1 Introduction 150 7.2 151 7.3 155 References 156 List of symbols 157 8 Risk attribution in a global country-sector model 159 Alan Scowcroft and James Sefton 8.1 Introduction 159 8.2 Recent trends in the ‘globalization’ of equity markets 161 8.2.1 ‘Home bias’ 162 8.2.2 The rise and rise of the multinational corporation 165 8.2.3 Increases in market concentration 167 8.3 Modelling country and sector risk 170 8.4 The estimated country and sector indices 176 viii Contents 8.5 Stock and portfolio risk attribution 181 8.6 Conclusions 188 8.7 Further issues and applications 189 8.7.1 Accounting for currency risk 189 8.7.2 Additional applications for this research 190 References 190 Appendix A: A detailed description of the identifying restrictions 193 Appendix B: The optimization algorithm 197 Appendix C: Getting the hedge right 199 9 Predictability of fund of hedge fund returns using DynaPorte 202 Greg N. Gregoriou and Fabrice Rouah 9.1 Introduction 202 9.2 Literature review 203 9.3 Methodology and data 204 9.4 Empirical results 204 9.5 Discussion 205 9.6 Conclusion 207 References 207 10 Estimating a combined linear factor model 210 Alvin L. Stroyny 10.1 Introduction 210 10.2 A combined linear factor model 211 10.3 An extended model 213 10.4 Model estimation 214 10.5 Conditional maximization 216 10.6 Heterogeneous errors 217 10.7 Estimating the extended model 218 10.8 Discussion 220 10.9 Some simulation evidence 221 10.10 Model extensions 222 10.11 Conclusion 223 References 224 11 Attributing investment risk with a factor analytic model 226 Dr T. Wilding 11.1 Introduction 226 11.2 The case for factor analytic models 227 11.2.1 Types of linear factor model 227 11.2.2 Estimation issues 228 11.3 Attributing investment risk with a factor analytic model 229 11.3.1 Which attributes can we consider? 230 11.4 Valuation attributes 231 11.4.1 Which attributes should we consider? 231 Contents ix 11.4.2 Attributing risk with valuation attributes 236 11.5 Category attributes 237 11.5.1 Which categories should we consider? 239 11.5.2 Attributing risk with categories 240 11.6 Sensitivities to macroeconomic time series 241 11.6.1 Which time series should we consider? 241 11.6.2 Attributing risk with macroeconomic time series 241 11.7 Reporting risk – relative marginals 242 11.7.1 Case study: Analysis of a UK portfolio 244 11.8 Conclusion 245 References 246 Appendix 247 12 Making covariance-based portfolio risk models sensitive to the rate at which markets reflect new information 249 Dan diBartolomeo and Sandy Warrick, CFA 12.1 Introduction 249 12.2 Review 250 12.3 Discussion 253 12.4 The model 254 12.5 A few examples 257 12.6 Conclusions 259 References 259 13 Decomposing factor exposure for equity portfolios 262 David Tien, Paul Pfleiderer, Robert Maxim and Terry Marsh 13.1 Introduction 262 13.2 Risk decomposition: cross-sectional characteristics 264 13.3 Decomposition and misspecification in the cross-sectional model: a simple example 269 13.3.1 Industry classification projected onto factor exposures 269 13.3.2 Incorporating expected return information 270 13.4 Summary and discussion 273 References 274 Index 277