Abstract
Multi-factor risk models have been used in portfolio selection since the 1960s and early 1970s. The work of Barr Rosenberg led to the creation of Barra, the first major commercially available portfolio selection software. Rudd and Clasing brought Barra to the academic audience as professional managers were embracing it in industry. It is important to see how quantitative analysis was developed in investment research and analysis. Harry Markowitz, Bill Sharpe, Jan Mossin, and Jack Treynor pioneered capital market equilibrium and the creation and estimation of the Capital Asset Pricing Model. In the 1970s, multi-factor risk models were developed and estimated by Barr Rosenberg, Andrew Rudd, and their colleagues at Barra; John Blin and Steve Bender at APT; and Sebastian Ceria at Axioma. Multi-factor models pushed out the efficient frontier relative to a single factor risk model. McKinley Capital Management developed its Horse Race risk model philosophy to assess the most effective risk models for maximizing the Geometric Means, Sharpe Ratios, and Information Ratios. Several special issues of practitioner-oriented applied investment research featured the McKinley Horse Race results and reported that properly constructed multi-factor risk models created portfolios in which tilt variables that have statistically significant ICs produced statistically significant Active Returns. One must remove the market effect and extra-market covariances to properly estimate the contribution of a variable to the creation of efficient portfolios. APT and Axioma have reigned supreme in those tests. The integration of Axioma and AAF and ITG transactions costs on the FactSet Investment have led MCM to implement an Axioma portfolio.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
- 2.
Dow and DuPont merged in 2015, whereas the correlation comments were made in 2007.
- 3.
The reader is referred to the BARRA (US-E3) United States Equity, Version 3, Risk Model Handbook. Guerard and Mark (2003 and 2020) and Miller, Xu, and Guerard (2014) used the US-E3 model in their analysis. The reader is also referred to Rosenberg and Marathe (1979), Rudd and Rosenberg (1980), Rudd and Clasing (1982), and Grinold and Kahn (1999).
- 4.
The reader is referred to Bruce and Epstein (1994) for a collection of published earnings forecasting research during the 1968–1992 time period, and to Brown (1998 and 2008) for a comprehensive annotated bibliography of this research.
- 5.
The forecast earnings per share for the 1-year-ahead and 2-year-ahead periods, FEP1 and FEP2, offer negative, but statistically insignificant, asset selection. The total active returns are positive, and not statistically significant. The asset selection is negative, because the FEP variables have positive and statistically significant loadings on the risk indices – particularly the earnings yield index. The factor loading of the FEP variables on the earnings yield risk index is not unexpected given that the earnings yield factor index in the US-E3 includes the forecast earnings-to-price variable. Thus, there is no multiple-factor model benefit to the FEP variables. The monthly revision variables, the RV variables, offer no statistically significant total active returns, or asset selection abilities. The breadth variables, BR, produce statistically significant total active returns and asset selection, despite statistically significant risk index loadings. The breadth variable loads on the earnings yield and growth risk indices. Taking a closer look at the BR1 factor risk index loading, the BR1 variable leads a portfolio manager to have a positive average active exposure to the earnings yield index, which incorporates the analyst predicted earnings-to-price and historical earnings-to-price measures. The BR1 tilt has a negative and statistically significant average exposure to size, nonlinearity, and the cube of normalized market capitalization. This result is consistent with analyst revisions being more effective in smaller-capitalized securities.
- 6.
Jose Menchero updated the Barra Global Equity Model (GEM) in 2010, see Menchero et al. (2010). Menchero, Morozov, and Shepard estimated the GEM2 Model, which had eight style factors: volatility, momentum, size, value, growth, nonlinear size, liquidity, and leverage. The original Global Equity Model (GEM) Model that was in place when John Guerard became Director of Quantitative Research at McKinley Capital Management, in 2005, had four style factors: size (the logarithm of market capitalization), success (composed of relative strength), value (composed of earnings to price, book to price, dividend yield, and analyst predicted earnings to price), and variability in markets (historical sigma). Menchero et al. (2010) created three sets of factor portfolios. First, simple factor returns, with factors standardized to have cap-weighted mean 0 and standard deviation 1, producing factor-replicating portfolios that go long stocks with positive factor exposures and short stocks with negative factor exposures. Simple factor portfolios can have significant exposures to other factors. Second, pure factor exposures in which every stock has unit exposure to the World factor and exposures of 0 and 1 to country and industries. Pure factor-replicating portfolios go long stocks with positive factor exposures and short stocks with negative factor exposures and have zero exposures to other style factors. The third set of factor portfolios are optimized factor portfolios which have unit exposure to the particular factor and zero exposures to all other factors. The GEM2 momentum factor enhanced wealth in simple and pure factor models portfolios, which dominated the optimized momentum factor portfolio for the 1996–2010 time period. Optimized factor portfolios for size, nonlinear size, liquidity, and leverage underperformed the simple and pure factor portfolios, implying that the impact of these variables weakened owing to country, industry, and style factor interactions. See Menchero et al. (2010) in Guerard (2010) and Lee et al. (2013), Chap. 25, written by Menchero, Morozov, and Guerard.
- 7.
Guerard and Schwartz (2007), Chap. 15, identified a lambda of 200 as the Sharpe-Ratio-maximizing lambda. Further support can be found in Guerard, Rachev, and Shao (2013), reprinted in Guerard and Ziemba (2020).
- 8.
A very important result of APT portfolio modeling is reported in Guerard, Krauklis, and Kumar (2012) for the application of mean-variance, enhanced index tracking and Tracking-at-Risk optimization techniques for the 1997–2009 time period. We can optimize the US stocks to create portfolios in the Russell 3000 universe for the 12-year backtest period, 1997–2009, where the absolute value of the stock weight cannot deviate by more than 2%. Guerard, Krauklis, and Kumar labeled these as (1) an EIT portfolio or (2) an EAW2 (equal active weight of 2%) portfolio. For a lambda of 200, 99 stocks are needed in the mean-variance portfolios and 108 stocks in the EAW2 portfolios. For a lambda of 10, we need 159 stocks in the mean-variance portfolios and 158 stocks in the EAW2 portfolios. An index-hugging portfolio – one designed by a manager to avoid taking active bets – would use a lambda of 10. An even less aggressive manager would need 161 stocks in an EAW1 portfolio for a lambda of 10, whereas a manager using a Sharpe-Ratio-maximizing lambda of 100 would need 131 stocks.
- 9.
Guerard brought the Blin and Bender APT system to MCM in August 2005. The APT system produced real-time portfolios that generated statistically significant Active Returns and Specific Returns. See Guerard, Deng, Gillam, Markowitz, and Xu (2019). The real-time Specific Returns exceeded 300 basis points, very consistent with the model’s backtests. See Guerard and Markowitz (2018). Full disclosure: Guerard worked with APT as a client from 1988 until early 2008, when Blin and Bender sold the company to SunGard.
- 10.
Axioma Risk Model Handbook, January 2010.
- 11.
In 2017, MCM ran its third Horse Race using Axioma with integrated ITG transactions costs, in the MSCI Non-US (XUS), Global (GL), and Emerging Markets (EM) universes. The XUS, GL, and EM reported results show much higher Active and Specific Returns in non-US markets than in the US market. International markets are more inefficient than the US markets, as was reported in Guerard and Mark (2020).
References
Barillas, F., and J. Shanken. 2018. Comparing asset pricing models. Journal of Finance 73: 715–755.
Baumol, W.J. 1963. An expected gain-confidence limit criteria for portfolio selection. Management Science 10: 174–182.
Black, F., M.C. Jensen, and M. Scholes. 1972. The capital asset pricing model: Some empirical tests. In Studies in the theory of capital markets, ed. M. Jensen. New York: Praeger.
Blin, J.M., S. Bender, and J.B. Guerard Jr. 1997. Earnings forecasts, revisions and momentum in the estimation of efficient market-neutral Japanese and U.S. portfolios. In Research in finance, ed. A. Chen, vol. 15: 93–114. Greenwich: JAI Press.
Bloch, M., J.B. Guerard Jr., H.M. Markowitz, P. Todd, and G. Xu. 1993. A comparison of some aspects of the U.S. and Japanese equity markets. Japan and the World Economy 5: 3–26.
Borkovec, M., I. Domowitz, B. Kiernan, and V. Serbin. 2010. Portfolio optimization and the cost of trading. Journal of Investing 19: 63–76.
Bruce, B., and Epstein, C. B. 1994. The Handbook of Corporate Earnings Analysis. Chicago: Probus Publishing Company.
Ceria, S., A. Saxena, and R.A. Stubbs. 2012. Factor alignment problems and quantitative portfolio management. The Journal of Portfolio Management 38(2): 29–43.
Cohen, K.J., and J.A. Pogue. 1967. An empirical evaluation of alternative portfolio-selection models. Journal of Business 40: 166–193.
Conner, G., and R.A. Korajczyk. 1995. The arbitrary theory and multifactor models of asset returns. In Handbooks in operations research and management science: Finance, ed. R.A. Jarrow, V. Malsimovic, and W.T. Ziemba. Amsterdam: North-Holland.
Connor, G., L. Goldberg, and R.A. Korajczyk. 2010. Portfolio risk analysis. Princeton: Princeton University Press.
Dhrymes, P.J., I. Friend, and N.B. Gültekin. 1984. A critical re-examination of the empirical evidence on the arbitrage pricing theory. Journal of Finance 39: 323–346.
Dhrymes, P.J., I. Friend, M.N. Gültekin, and N.B. Gültekin. 1985. New tests of the APT and their implications. Journal of Finance 40: 659–674.
Dimson, E. 1988. Stock Market Anomalies. Cambridge: Cambridge University Press.
Elton, E.J., and M.J. Gruber. 1970. Homogeneous groups and the testing of economic hypothesis. Journal of Financial and Quantitative Analysis 5: 581–602.
———. 1973. Estimating the dependence structure of share prices – Implications for portfolio selection. Journal of Finance 28: 1203–1232.
Elton, E.J., M.J. Gruber, and M.W. Padberg. 1978. Simple criteria for optimal portfolio selection: Tracing out the efficient frontier. Journal of Finance 33: 296–302.
———. 1979. Simple criteria for optimal portfolio selection: The multi-index case. In Portfolio theory, 25 years after: Essays in honor of Harry Markowitz, ed. E.J. Elton and M.J. Gruber. Amsterdam: North-Holland.
Elton, E.J., M.J. Gruber, S.J. Brown, and W.N. Goetzman. 2007. Modern portfolio theory and investment analysis. 7th ed. New York: Wiley.
Fama, E.F. 1976. Foundations of finance. New York: Basic Books.
Fama, E.F. and K.R. French. 1992. Cross-sectional variation in expected stock returns. Journal of Finance 47: 427–465.
———. 1995. Size and the book-to-market factors in earnings and returns. Journal of Finance 50: 131–155.
———. 2008. Dissecting anomalies. Journal of Finance, 63: 1653–1678.
Fama, E.F., and J.D. MacBeth. 1973. Risk, return, and equilibrium: Empirical tests. Journal of Political Economy 81: 607–636.
Farrell, J.L., Jr. 1974. Analyzing covariance of returns to determine homogeneous stock grou**s. Journal of Business 47: 186–207.
———. 1997. Portfolio management: Theory and applications. New York: McGraw-Hill/Irwin.
Grinold, R., and R. Kahn. 1999. Active portfolio management. New York: McGraw-Hill/Irwin.
Guerard, J.B., Jr., and A. Mark. 2003. The optimization of efficient portfolios: The case for a quadratic R and D term. In Research in finance, ed. A. Chen. 20: 213–247. Greenwich: JAI Press.
Guerard, J.B., Jr., R.A. Gillam, H.M. Markowitz, G. Xu, S. Deng, and Z. Wang. 2018. Data mining corrections testing in Chinese stocks. Interfaces 48: 108–120.
Guerard, J.B., Jr., G. Xu, and Z. Wang. 2019. Portfolio and investment analysis with SAS@: Financial modeling techniques for optimization. Cary: SAS Press.
Guerard, J.B., Jr. 2012. Global earnings forecasting efficiency. In Research in finance, ed. J. Kenningser. Bingley: Emerald Publishing Company.
Guerard, J.B., Jr., and A. Mark. 2020. Earnings forecasts and revisions, price momentum, and fundamental data: Further explorations of financial anomalies. In Handbook of financial econometrics, ed. C.F. Lee and J.C. Lee, vol. 2. Singapore: World Scientific Handbook in Financial Economics.
Guerard, J.B., Jr., and E. Schwartz. 2007. Quantitative corporate finance. New York: Kluwer Academic Publishers.
Guerard, J.B., Jr., and W.T. Ziemba. 2020. Handbook of applied investment research. Singapore: World Scientific Publishing.
Guerard, J.B., Jr., M. Gültekin, and B.K. Stone. 1997. The role of fundamental data and analysts’ earnings breadth, forecasts, and revisions in the creation of efficient portfolios. In Research in finance, ed. A. Chen, vol. 15. Greenwich: JAI Press.
Guerard, J.B., Jr., G. Xu, and M.N. Gültekin. 2012. Investing with momentum: The past, present, and future. Journal of Investing 21: 68–80.
Guerard, J.B., Jr., R.T. Rachev, and B. Shao. 2013. Efficient global portfolios: Big data and investment universes. IBM Journal of Research and Development 57: 11:1–11:11.
Guerard, J.B., Jr., H.M. Markowitz, and G. Xu. 2014. The role of effective corporate decisions in the creation of efficient portfolios. IBM Journal of Research and Development 586: 1–11.
———. 2015. Earnings forecasting in a global stock selection model and efficient portfolio construction and management. International Journal of Forecasting 31: 550–560.
Guerard, J.B., Jr., H.M. Markowitz, G. Xu, and Z. Wang. 2018. Global portfolio construction with emphasis on conflicting corporate strategies to maximize stockholder wealth. Annals of Operations Research 267: 203–219.
Hirshleifer, D. 2001. Investor psychology and asset pricing. Journal of Finance 64: 1533–1597.
Jacobs, B.I., and K. Levy. 2017. Equity management: The art and science of modern quantitative investing. 2nd ed. New York: McGraw-Hill.
Jensen, M. 1972. Studies in the theory of capital markets. New York: Praeger.
King, B.F. 1966. Market and industry factors in stock price behavior. Journal of Business 39: 139–191.
Lee, C.F., J. Finnerty, J. Lee, A.C. Lee, and D. Wort. 2013. Security analysis, portfolio management, and financial derivatives. Singapore: World Scientific Publishing Company.
Markowitz, H.M. 1959. Portfolio selection: Efficient diversification of investment, Cowles foundation monograph no. 16. New York: Wiley.
———. 2000. Mean-variance analysis in portfolio choice and capital markets. New Hope: Frank J. Fabozzi Associates.
Menchero, and Z. Nagy. 2017. Performance of earnings yield and momentum factors in US and International Equity Markets. In Portfolio construction, measurement, and efficiency: Essays in Honor of Jack Treynor, ed. J.B. Guerard. New York: Springer.
Miller, W., G. Xu, and J.B. Guerard Jr. 2014. Portfolio construction and management in the BARRA aegis system: A case study using the USER data. Journal of Investing 23: 111–120.
Mossin, J. 1966. Equilibrium in a capital asset market. Econometrica 34: 768–783.
———. 1973. Theory of financial markets. Englewood Cliffs: Prentice-Hall.
Reinganum, M. 1981. The arbitrage pricing theory: Some empirical tests. Journal of Finance 36: 313–321.
Roll, R., and S.A. Ross. 1984. The arbitrage pricing theory approach to strategic portfolio planning. Financial Analysts Journal 40 (3): 14–26.
Rosenberg, B. 1974. Extra-market components of covariance in security returns. Journal of Financial and Quantitative Analysis 9: 263–274.
———. 1976. Security appraisal and unsystematic risk in institutional investment. Proceedings of the seminar for research in security prices. University of Chicago.
Rosenberg, B., and V. Marathe. 1979. Tests of capital asset pricing hypotheses. In Research in finance, ed. H. Levy. Greenwich, Ct: JAI Press. 1: 115–224.
Rosenberg, B., and W. McKibben. 1973. The prediction of systematic and unsystematic risk in common stocks. Journal of Financial and Quantitative Analysis 8: 317–333.
Rosenberg, B., and A. Rudd. 1977. The yield/beta/residual risk tradeoff, Working paper no. 66. Research program in finance. Berkeley: Institute of Business and Economic Research, University of California.
Rosenberg, B., M. Hoaglet, V. Marathe, and W. McKibben. 1975. Components of covariance in security returns, Working paper no. 13. Research program in finance. Berkeley: Institute of Business and Economic Research, University of California.
Ross, S.A. 1976. The arbitrage theory of capital asset pricing. Journal of Economic Theory 13: 341–360.
Ross, S.A., and R. Roll. 1980. An empirical investigation of the arbitrage pricing theory. Journal of Finance 35: 1071–1103.
Rudd, A., and H.K. Clasing. 1982. Modern portfolio theory: The principles of investment management. Homewood: Dow Jones-Irwin.
Rudd, A., and B. Rosenberg. 1979. Realistic portfolio optimization. In Portfolio theory, 25 years after, ed. E. Elton and M. Gruber. Amsterdam: North-Holland.
———. 1980. The “Market Model” in investment management. Journal of Finance 35: 597–607.
Saxena, A., and R.A. Stubbs. 2012. An empirical case study of factor alignment using the USER model. Journal of Investing 21: 25–44.
———. 2013. Alpha alignment factor: A solution to the underestimation of risk for optimized active portfolios. The Journal of Risk 153: 3–37.
———. 2015. Augmented risk models to mitigate factor alignment problems. Journal of Investment Management 13 (3): 57–79.
Sharpe, W.F. 1963. A simplified model for portfolio analysis. Management Science 9: 277–293.
———. 1964. Capital asset prices: A theory of market equilibrium under conditions of risk. Journal of Finance 19: 425–442.
———. 1971. A linear programming approximation for the general portfolio analysis problem. Journal of Financial and Quantitative Analysis 6:1263–1275.
———. 2012. William F. Sharpe: Selected works. Singapore: World Scientific Publishing Company.
Stone, B.K. 1970. Risk, return, and equilibrium: A general single-period theory of asset selection and capital market equilibrium. Cambridge, MA: MIT Press.
———. 1973. A linear programming formulation of the general portfolio selection problem. Journal of Financial and Quantitative Analysis 8: 621–636.
———. 1974. Systematic interest-rate risk in a two-index model of returns. Journal of Financial and Quantitative Analysis 9: 709–721.
Ziemba, W.T. 1993. Invest Japan. Chicago: Probus Publishing Co.
Acknowledgments
The authors would like to thank research conversations with Doug Martin, Bijan Beheshti, and Sebastian Ceria. Any errors remaining are the sole responsibility of the authors.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendices
Appendix 1: US-E3 Descriptor Definitions
This appendix gives detailed definitions of the descriptors that underlie the risk indices in US-E3. The method of combining these descriptors into risk indices is proprietary to Barra.
1. Volatility
-
(i)
BTSG: Beta times sigma
This is computed as \( \sqrt{\beta {\sigma}_{\varepsilon }} \), where β is the historical beta and σε is the historical residual standard deviation. If β is negative, then the descriptor is set equal to zero.
-
(ii)
DASTD: Daily standard deviation
This is computed as:
where rt is the return over day t, wt is the weight for day t, T is the number of days of historical returns data used to compute this descriptor (we set this to 65 days), and Ndays is the number of trading days in a month (we set this to 23).
-
(iii)
HILO: Ratio of high price to low price over the last month
This is calculated as:
where PH and PL are the maximum price and minimum price attained over the last 1 month.
-
(iv)
LPRI: Log of stock price
This is the log of the stock price at the end of last month.
-
(v)
CMRA: Cumulative range
Let Zt be defined as follows:
where ri,s is the return on stock I in month s, and rf,s is the risk-free rate for month s. In other words, Zt is the cumulative return of the stock over the risk-free rate at the end of month t. Define Zmax and Zmin as the maximum and minimum values of Zt over the last 12 months. CMRA is computed as:
-
(vi)
VOLBT: Sensitivity of changes in trading volume to changes in aggregate trading volume
This may be estimated by the following regression:
where ΔVi,t is the change in share volume of stock I from week t-1 to week t, Ni,t is the average number of shares outstanding for stock I at the beginning of week t-1 and week t, ΔVM,t is the change in volume on the aggregate market from week t-1 to week t, and NM,t is the average number of shares outstanding for the aggregate market at the beginning of week t-1 and week t.
-
(vii)
SERDP: Serial dependence
This measure is designed to capture serial dependence in residuals from the market model regressions. It is computed as follows:
where et is the residual from the market model regression in month t, and T is the number of months over which this regression is run (typically, T = 60 months).
-
(viii)
OPSTD: Option-implied standard deviation
This descriptor is computed as the implied standard deviation from the Black-Scholes option pricing formula using the price on the closest to at-the-money call option that trades on the underlying stock.
2. Momentum
-
(i)
RSTR: Relative strength
This is computed as the cumulative excess return (using continuously compounded monthly returns) over the last 12 months – i.e.,
$$ \mathrm{RSTR}\stackrel{\longrightarrow}{\leftarrow} ={\sum}_{t=1}^T\log \left(1+{r}_{\mathrm{i},\mathrm{t}}\right)-{\sum}_{t=1}^T\log \left(1+{r}_{\mathrm{f},\mathrm{t}}\right) $$
where ri,t is the arithmetic return of the stock in month I, and rf,t is the arithmetic risk-free rate for month i. This measure is usually computed over the last 1 year – i.e., T is set equal to 12 months.
-
(ii)
HALPHA: Historical alpha
This descriptor is equal to the alpha term (i.e., the intercept term) from a 60-month regression of the stock’s excess returns on the S&P 500 excess returns.
3. Size
-
(i)
LNCAP: Log of market capitalization
This descriptor is computed as the log of the market capitalization of equity (price times number of shares outstanding) for the company.
4. Size nonlinearity
-
(i)
LCAPCB: Cube of the log of market capitalization
This risk index is computed as the cube of the normalized log of market capitalization.
5. Trading activity
-
i)
STOA: Share turnover over the last year
STOA is the annualized share turnover rate using data from the last 12 months – i.e., it is equal to Vann / \( \overline{N} \)out, where Vann is the total trading volume (in number of shares) over the last 12 months and \( \overline{N} \)out is the average number of shares outstanding over the previous 12 months (i.e., it is equal to the average value of the number of shares outstanding at the beginning of each month over the previous 12 months).
-
ii)
STOQ: Share turnover over the last quarter
This is computed as the annualized share turnover rate using data from the most recent quarter. Let Vq be the total trading volume (in number of shares) over the most recent quarter and let \( \overline{N} \)out be the average number of shares outstanding over the period (i.e., \( \overline{N} \)out is equal to the average value of the number of shares outstanding at the beginning of each month over the previous 3 months). Then, STOQ is computed as 4Vq / \( \overline{N} \)out.
-
iii)
STOM: Share turnover over the last month
This is computed as the share turnover rate using data from the most recent month – i.e., it is equal to the number of shares traded last month divided by the number of shares outstanding at the beginning of the month.
-
iv)
STO5: Share turnover over the last 5 years
This is equal to the annualized share turnover rate using data from the last 60 months. In symbols, STO5 is given by:
where Vs is equal to the total trading volume in months, and \( \overline{N} \)out is the average number of shares outstanding over the last 60 months.
-
v)
FSPLIT: Indicator for forward split
This descriptor is a 0–1 indicator variable to capture the occurrence of forward splits in the company’s stock over the last 2 years.
-
vi)
VLVR: Volume to variance
This measure is calculated as follows:
where Vs equals the number of shares traded in month s, Ps is the closing price of the stock at the end of month s, and σε is the estimated residual standard deviation. The sum in the numerator is computed over the last 12 months.
6. Growth
-
(i)
PAYO: Payout ratio over 5 years
This measure is computed as follows:
where Dt is the aggregate dividend paid out in year t and Et is the total earnings available for common shareholders in year t. This descriptor is computed using the last 5 years of data on dividends and earnings.
-
(ii)
VCAP: Variability in capital structure
This descriptor is measured as follows:
where Nt-1 is the number of shares outstanding at the end of time t-1; Pt-1 is the price per share at the end of time t-1; LDt-1 is the book value of long-term debt at the end of time period t-1; PEt-1 is the book value of preferred equity at the end of time period t-1; and CET + LDT + PET are the book values of common equity, long-term debt, and preferred equity as of the most recent fiscal year.
-
(iii)
AGRO: Growth rate in total assets
To compute this descriptor, the following regression is run:
where TAit is the total assets of the company as of the end of year t, and the regression is run for the period = 1, …, 5. AGRO is computed as follows:
where the denominator average is computed over all the data used in the regression.
-
(iv)
EGRO: Earnings growth rate over last 5 years
First, the following regression is run:
where EPSt is the earnings per share for year t. This regression is run for the period t = 1, …, 5. EGRO is computed as follows:
-
(v)
EGIBS: Analyst-predicted earnings growth
This is computed as follows:
where EARN is a weighted average of the median earnings predictions by analysts for the current year and next year, and EPS is the sum of the four most recent quarterly earnings per share.
-
(vi)
DELE: Recent earnings change
This is a measure of recent earnings growth and is measured as follows:
where EPSt is the earnings per share for the most recent year, and EPSt-1 is the earnings per share for the previous year. We set this to missing if the denominator is nonpositive.
7. Earnings Yield
-
(i)
EPIBS: Analyst-predicted earnings-to-price
This is computed as the weighted average of analysts’ median predicted earnings for the current fiscal year and next fiscal year divided by the most recent price.
-
(ii)
ETOP: Trailing annual earnings-to-price
This is computed as the sum of the four most recent quarterly earnings per share divided by the most recent price.
-
(iii)
ETP5: Historical earnings-to-price
This is computed as follows:
where EPSt is equal to the earnings per share over year t, and Pt is equal to the closing price per share at the end of year t.
8. Value
-
(i)
BTOP: Book-to-price ratio
This is the book value of common equity as of the most recent fiscal year-end divided by the most recent value of the market capitalization of the equity.
9. Earnings Variability
-
(i)
VERN: Variability in earnings
This measure is computed as follows:
where Et is the earnings at time t(t = 1,…,5), and Ē is the average earnings over the last 5 years. VERN is the coefficient of variation of earnings.
-
(ii)
VFLO: Variability in cash flows
This measure is computed as the coefficient of variation of cash flow using data over the last 5 years – i.e., it is computed in an identical manner to VERN, with cash flow being used in place of earnings. Cash flow is computed as earnings plus depreciation plus deferred taxes.
-
(iii)
EXTE: Extraordinary items in earnings
This is computed as follows:
where EXt is the value of extraordinary items and discontinued operations, NRIt is the value of nonoperating income, and Et is the earnings available to common before extraordinary items. The descriptor uses data over the last 5 years.
-
(iv)
SPIBS: Standard deviation of analysts’ prediction to price
This is computed as the weighted average of the standard deviation of IBES analysts’ forecasts of the firm’s earnings per share for the current fiscal year and next fiscal year divided by the most recent price.
10. Leverage
-
(i)
MLEV: Market leverage
This measure is computed as follows:
where MEt is the market value of common equity, PEt is the book value of preferred equity, and LDt is the book value of long-term debt. The value of preferred equity and long-term debt are as of the end of the most recent fiscal year. The market value of equity is computed using the most recent month’s closing price of the stock.
-
(ii)
BLEV: Book leverage
This measure is computed as follows:
where CEQt is the book value of common equity, PEt is the book value of preferred equity, and LDt is the book value of the long-term debt. All values are as of the end of the most recent fiscal year.
-
(iii)
DTOA: Debt-to-assets ratio
This ratio is computed as follows:
where LDt is the book value of long-term debt, DCLt is the value of debt in current liabilities, and TAt is the book value of total assets. All values are as of the end of the most recent fiscal year.
-
(iv)
SNRRT: Senior debt rating
This descriptor is constructed as a multi-level indicator variable of the debt rating of a company.
11. Currency Sensitivity
-
(i)
CURSENS: Exposure to foreign currencies
To construct this descriptor, the following regression is run:
where rit is the excess return on the stock and rmt is the excess return on the S&P 500 Index. Let εit denote the residual returns from this regression. These residual returns are in turn regressed against the contemporaneous and lagged returns on a basket of foreign currencies, as follows:
where εit is the residual return on stock I, (FX)t is the return on an index of foreign currencies over month t, (FX)t-1 is the return on the same index of foreign currencies over month t-1, and (FX)t-2 is the return on the same index over month t-2. The risk index is computed as the sum of the slope coefficients γi1, γi2, and γi3 – i.e., CURSENS = γi1 + γi2 + γi3.
12. Dividend Yield
-
(i)
P_DYLD: Predicted dividend yield
This descriptor uses the last four quarterly dividends paid out by the company along with the returns on the company’s stock and future dividend announcements made by the company to come up with a Barra-predicted dividend yield.
13. Non-estimation Universe Indicator
-
(i)
NONESTU: Indicator for firms outside US-E3 estimation universe
This is a 0–1 indicator variable: It is equal to 0 if the company is in the BARRA estimation universe and equal to 1 if the company is outside the BARRA estimation universe.
Appendix 2 Barra US-E4, GEM4 Earnings Yields and Momentum Strategies
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this entry
Cite this entry
Blin, J., Guerard, J., Mark, A. (2022). A History of Commercially Available Risk Models. In: Lee, CF., Lee, A.C. (eds) Encyclopedia of Finance. Springer, Cham. https://doi.org/10.1007/978-3-030-91231-4_99
Download citation
DOI: https://doi.org/10.1007/978-3-030-91231-4_99
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-91230-7
Online ISBN: 978-3-030-91231-4
eBook Packages: Economics and FinanceReference Module Humanities and Social SciencesReference Module Business, Economics and Social Sciences