Corneille, Sperna Weiland, and Zürcher Win European QuantAwards 2014

Spectacular improvements in computing technology, data, and statistical machinery in recent years have stimulated impressive leaps of thought and techniques in quantitative finance. An at times reluctant investment industry worldwide has embraced and adapted many of these new ideas, which are now ubiquitous in the modern asset management industry: passive indexation, active quantitative investment, factor investing, risk measurement and management, derivatives, hedge funds, complex trading strategies, and more all rely on the new quant techniques.

Qualitative investing styles, with fund managers handpicking stocks by focusing on intangibles, are the traditional counterpoint to quantitative investing. But even the most purist qualitative investors, in an era exploding with quickly processed information, are turning to quantitative techniques to control and isolate the information needed for their more intuitive approach. Pioneers of investment such as Ben Graham and John Maynard Keynes were each skilled measurers as much as superlative interpreters of the resultant data. So while the future of the investment industry likely holds a quickening pace of innovation and a cross-fertilization among quant and other approaches, efforts to research and develop applied quantitative strategies should be applauded.

At the recent European QuantAwards 2014 competition, in association with State Street Global Advisors, many entrants considered the current state of quantitative investing. Here is a summary of the three winning papers written by their student authors, René-Jean Corneille, ESCP Europe, France; Rob Sperna Weiland, VU University Amsterdam, the Netherlands; and Ulrik Zürcher, Maastricht University, the Netherlands.

First Prize: René-Jean Corneille. “Exploring Macroeconomic Sensitivities of Style Risk Premia: A Bayesian Time Series Approach to Optimal Investing.”

We introduce a new weighting scheme to optimally allocate assets in a global-macro strategy using style risk premia strategies. This framework provides a twist to traditional asset classification. The strategy is based on a four-state Markov-switching modeling of macroeconomic data.

The basic building blocks of this new portfolio are systematic risk premia set up using several asset classes: commodities, interest rates, equities, and exchange rates in a number of geographical zones, including the United States, the United Kingdom, the eurozone, Japan, and emerging countries. Bayesian inference is employed to estimate the model parameters which are used to compute the optimal weights of the systematic strategies in the overall portfolio.

More precisely, the weights are a function of the regime probability transition matrix and the sensitivities of each style risk premia with relation to the different regimes. Regimes tested are the following: growth, crisis, recovery, and slow-down (or market stress). In this way, the strategy requires views for every risk premia with relation to each regime, which allows greater flexibility and consideration of the possible downside risk premia.

We find that combining several risk premia in a portfolio achieves better diversification. Our optimal strategy shows very promising performance against traditional benchmarks: the probability transition weights have a smoothing effect, reducing the portfolio volatility.

Second Prize: Rob Sperna Weiland. “Liquidity Risk in the Sovereign Credit Default Swap Market.”

In this paper, which is an adaptation of a study conducted at Rabobank, we investigate the use of sovereign credit default swap (CDS) premia in order to estimate sovereign default probabilities. We conjecture that liquidity risk is highly priced into these premia and that we therefore need to measure and account for this distorting effect in order to get uncontaminated estimates of the default probabilities.

We introduce an intensity-based model that allows for a country-specific analysis and induces a natural decomposition of the CDS premia into a credit part and a liquidity part. We test our model on Brazilian and Turkish CDS data and we show that liquidity risk is indeed highly priced into the CDS premia. Our default probability estimates are close to Rabobank’s internal estimates, which boosts the confidence we have in our proposed methodology.

Third Prize: Ulrik Zürcher. “The Effect of Interest Rates on Equity Markets that Allows Share Repurchases: A New Paradigm.”

A common approach to gauge the “exuberance” in stock markets is to compare historical valuation multiples to current levels. This methodology is unhelpful in low-yielding environments, such as the present day, which are characterized by high multiples and low interest rates, as you are effectively comparing apples and oranges — or different ex-ante yield regimes. Most academics and practitioners agree that valuation multiples (in aggregate) should be relatively higher when interest rates are lower and vice versa. Nevertheless, the exact dynamics of this relationship are poorly understood.

My research demonstrates that aggregate P/E ratios  are co-integrated with average corporate after-tax interest rates. The theoretical underpinnings of this result are that it can be demonstrated that reaching this equilibrium will maximize EPS. This motivates companies to actively pursue equilibrium by engineering their capital structure. Furthermore, based on these findings, I built an error correction model that proves superior to many current modeling attempts of the P/E ratio found in academia.

Overall, this discussion presents a view, which is strongly supported by the data, on the dynamics between interest rates and stock market fundamentals — given that companies can freely manipulate their capital structure by, for example, buying back shares.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Photo credit: ©iStockphoto.com/Meriel Jane Waissman