This Market Insight presents the results of an annual backtesting study, using RiskManager, applied to four standard risk models.  The study includes fixed income and equity portfolios during the 2013 calendar year.  While the first half of 2013 was quiet, volatility increased after June 2013. For fixed income portfolios, comparing ex-ante risk forecasts with ex-post returns, the more reactive models showed some underestimation of risk in the turbulent period, while the more stable...

From quantitative easing to talk of tapering, a recurrent theme of 2013 among investors has been concern over rising interest rates. There are many ways to design a rising interest rate stress test,though basing this on history alone is a challenge. Historically, interest rates have risen for numerous reasons: central bank actions, inflation, flight to quality, and so on. The commonality between past regimes of rising interest rates is not obvious. In order to simplify and use a direct link...

It has been common practice for over a decade for banks to make public risk disclosures. The detail provided in these disclosures varies across banks, which makes comparison difficult. In this paper, we propose a set of standards in order to analyze these disclosures. In particular, we compare a sample of bank risk disclosures with our set of standards, identifying banks whose risk appears to have moved by enough to suggest a degree of active management.

A predictive stress test based on historical data can be a valuable tool in contemplating large market shocks. When the degree of extrapolation becomes excessive, or the prediction is based on a tenuous historical correlation, the predictive stress test can lead to unintuitive results that do not conform to reasonable market and economic expectations. This Research Insight offers best practices for users of RiskManager and BarraOne who want to design stress tests with correlated factors;...

This white paper presents model backtesting results, using RiskManager, for a number of standard risk models applied to fixed income and equity portfolios during the period December 2011 to November 2012.  This is a follow up to a similar exercise published for the 2011 calendar year.  Compared to 2011, this past year has been less dramatic, with all models producing relatively stable forecasts.  In terms of risk forecast performance, comparing ex-ante forecasts with ex-post...

In this paper, we investigate whether the new Barra Europe Stochastic Factor Model (EURS1) performs differently in portfolio construction when compared to other statistical models commonly used by investors.  In order to explore this, we built a typical principal component analysis (PCA) model and used both the PCA and EURS1 models to track two popular benchmarks using the Barra Open Optimizer.  In all cases, we found EURS1 to have the lowest tracking error, and the EURS1 slow model...

The enhancements introduced in the new Barra US Equity Model (USE4) are designed to improve risk forecasts for optimized portfolios. In this paper, we investigate the turnover and forecasting accuracy of optimized portfolios generated using USE4S, and compare them with those generated using the model's predecessor, USE3S. By testing various investment strategies, rebalancing on a daily or weekly basis, during the period 2007-2012, we found no systematic difference in turnover, and, in fact,...

Following a benign first quarter of 2012, investors have seen a breakdown in the Eurozone, which may cause a significant increase in risk aversion. Implied volatility is a common gauge of investor risk aversion and investor sentiment. In this report, we analyze factor returns in different volatility regimes and found that stocks with positive exposures to Size, Dividend Yield, Momentum and negative exposures to Residual Volatility and Leverage may have provided a hedge during rapid...

As measured by the VIX and VDAX, implied volatilities in both the US and Europe have fallen sharply since October 2011.  However, in the past ten years the declines in implied volatility have not been uniform: for example, when the dotcom bubble burst the VIX declined more rapidly than the VDAX.  During  the two most recent declines in volatility seen in 2009-2010 and 2011-2012, the US trended with global volatility while Europe decoupled from the global trend as it moved into...

The methodological enhancements introduced in the new Barra Global Equity Model (GEM3) are designed to improve risk forecasts for optimized portfolios. In this paper, we investigate the turnover and forecasting accuracy of optimized portfolios generated using GEM3, and compare them with those generated using the model’s predecessor, GEM2. By testing various investment strategies during the period 2007-2011, we found no systematic difference in turnover; in fact, the GEM3 risk forecasts...

From a modeling perspective, the heterogeneous character of the Asia Pacific region poses a particular challenge. The Barra ASE1 model takes this heterogeneity into account, striking an attractive balance between broad coverage and local detail. It provides regional market, industry, and style factors that help institutional investors compare drivers of risk on a pan-regional scale. Where necessary, the model adds local style and industry factors to enhance the explanatory power....