Financial Modeling of the Equity Market

From simple origins in the 1950s, portfolio theory has spawned a vast literature on asset and portfolio modeling. This book is a not-too-technical look at that modeling from the perspective of equity portfolio management. The target audience is portfolio managers or other investors who believe that markets are sufficiently inefficient to justify active management. The authors offer a whirlwind tour of modeling techniques that might prove useful in that endeavor. In this respect, the book fills a niche very similar to Grinold and Kahn (1999). The two books differ in that Grinold and Kahn is more practical. Fabozzi et al is more theoretical.

The authors describe numerous models. The sheer breadth of their coverage may be what makes the book most valuable. Models fall generally into two categories:

There are asset valuation models like CAPM and APT that can be used to construct optimal portfolios.

There are also time series or econometric models that can be used for constructing covariance matrices or other inputs for the first type of models.

   
 

Some models are covered in detail. Many are touched on only briefly, perhaps with a formula or two and a citation to the literature. In this regard, the book is like a survey of the literature. Or maybe I should describe it as a survey of available models. The authors are not particularly good at citing sources. They will cite a source, but that may not be the most relevant one. For example, they comment that mixed-distribution models were proposed in a 1999 paper. Those models have been around for decades, but you wouldn't know it from what the authors say.

Discussions can be vague or sloppy. The treatment of the Arbitrage Pricing Model is marred by cryptic footnotes cautioning that the discussion is not only simplified but technically incorrect!

Elsewhere, the authors claim that the concept of diversification depends on the central limit theorem. That theorem is all about normal distributions, and I don't see that normal distributions have anything to do with not placing all your eggs in one basket.

Contents

1. Introduction

2. Mean-variance analysis and modern portfolio theory

3. Transaction and trading costs

4. Applying the portfolio selection framework in practice

5. Incorporating higher moments and extreme risk measures

6. Mathematical and numerical optimization

7. Equity price models

8. Forecasting expected return and risk

9. Robust frameworks for estimation and portfolio allocation

10. Feedback and predictors in stock markets

11. Individual price processes : univariate models

12. Multivariate models

13. Model selection and its pitfalls

14. Estimation of regression models

15. Estimation of linear dynamic models

16. Estimation of hidden variable models

17. Model risk and its mitigation

App. A. Difference equations

App. B. Correlations, regressions, and copulas

A discussion of arbitrage is undermined by the authors equating an asset's payoff with its expected return.

In addition to such problems, there is a bigger problem, which really isn't the authors' fault. Implementing more advanced models, with all the data gathering, statistical analyses, coding and testing this entails, would require considerable time and resources. There is no evidence that doing so will contribute anything to portfolio returns. The whole book lies under a cloud called efficient markets theory. Even if you reject efficient markets theory, there is no means of identifying which, if any, of the models will add value. A book like this reminds me of the story of a farm boy digging through a huge pile of dung believing there must be a pony in there somewhere.

Despite its shortcomings, this book will be appealing to anyone with an interest in modeling portfolio risk and return. If you have read Grinold and Kahn and want to delve more deeply into modeling, this book will be an excellent jump-off point. It will present you with a vast array of ideas, so you can decide where to focus your studies. The book will appeal to other audiences as well. Discussions of covariance matrix construction will appeal to practitioners or theoreticians working with value-at-risk. Indeed, I think anyone with an interest in quantitative finance will find much amidst these pages that will stir their curiosity.[3/23/2006]

 

For related books, see sections:

Portfolio Management - General

Portfolio Management - Equities

Markets - Equities

Finance - Portfolio Theory

Finance - Financial Econometrics

Math - Time Series Analysis

Risk Management - Market Risk

 

Ads by Contingency Analysis.

Advertise on this site.

 

disclaimer

website: http://www.contingencyanalysis.com
books direct link: http://www.riskbook.com
copyright © Contingency Analysis, 1996 - current