What’s hot: a report on the Nomura Global Quantitative Equity Conference

The Nomura Quantitative Conference took place in early May of this year in London, and brought together a number of investment professionals, from asset managers to institutional investors. This conference is now a regular and important event attracting a large number of quantitative investment specialists and institutional investors.

I presented our most recent research papers on inter-temporal risk parity1,2, a strategy which changes allocation dynamically so that the ex-ante risk of the portfolio remains constant over time. In these two papers we have shown that investors can earn higher risk-adjusted returns if they manage the risk in their portfolios when investing in either an asset class such as equities or a risk-factor portfolio such as value or momentum. We demonstrate that the improvement of risk-adjusted returns arises from known properties of volatility of returns of financial assets and the relationship between volatility and returns.

Factor investing remains a hot topic with three presentations on the subject. The view at Allianz Global Investors (Allianz GI) seems to be in line with our research3. Like us, they think that smart-beta indexation is a poor approach to investing. Benedikt Henne from Allianz GI explained that different Smart Beta indices can have similar factor exposures and that it is difficult to manage the exposure to factors in portfolios of smart-beta indices. Moreover, they proposed an integrated solution, although they did not give details on how they put it together. Benedikt also presented a summary of the different explanations academics give for factor returns. He separated those based on assuming that factor returns are compensation for additional risk exposures from those which assume that factor returns are the result of anomalies or mispricing.

Wilma Groot from Robeco focused on this matter and showed that the returns from investing in value stocks and small-cap stocks do not seem to be compensation for additional risk exposures but are more likely the result of anomalies.

Finally, Florian Esterer from Bank J. Safra Sarasin discussed different methodologies for factor combination and portfolio construction with active factor exposures, acknowledging our recent paper³ on this topic but not discussing it.

There was also a panel discussion with David Buckle from Fidelity, Tim Wong from Man Group and Gerben De Zwart from APG on how to market quantitative funds. They highlighted the fact that quantitative managers tend to be poor at both marketing and keeping their clients aware of what occurs in their portfolios. Is this perhaps one of the reasons why quantitative investing is not stronger today? It was also pointed out that investors in quantitative funds need the reassurance that machines are not left to their own devices managing money quantitatively, but that there is someone supervising and making sure that everything runs smoothly.

On a related topic, Joseph Mezrich from Nomura talked about factor investing using machine learning techniques and in particular the CART4 algorithm. He finds it puzzling that these techniques, which are already used in a large number of industries, are still rarely used for managing money. This generated an intense debate with the audience thus demonstrating that the topic is still controversial even among the quantitative community.

Anthony Morris from Nomura touched upon a delicate subject, the recent paper 5 by David Bailey et al. in which the authors related financial mathematics to charlatanism, accusing quantitative managers of over-relying on backtesting and overfitting data. Anthony pointed out that the issues identified in David’s paper have long been known to the financial community and should not be dismissed out of hand. Daniel Kahneman, Nate Silver and Fisher Black have all previously considered the problem of overfitting. The problems highlighted in the David et al. paper can and are being tackled in different ways by quantitative managers. Moreover, according to Anthony, some of the solutions proposed by David et al. would not solve the underlying problems. In particular, as it was put by Stephen LeRoy in a 1973 paper5, data will never be enough. Anthony recognised perhaps an overuse of backtesting in David’s marketing materials from quantitative managers but made the point that without evidence of comprehensive back testing research would not be credible with investors.

On a quite different topic, Roni Israelov from AQR Capital presented an improved covered call strategy which is only exposed to equity risk premium and volatility premium, hedging all the equity market timing component away. According to Roni, this covered call strategy improves the risk-adjusted return over traditional covered call strategies, e.g. the strategy proposed by Chicago Board Options Exchange (CBOE), which did not remove the equity market timing component.

The conference was closed by Inigo Fraser Jenkins from Nomura who wrapped it up and briefly discussed the outlook for quantitative investing. He highlighted the fact that low active share quantitative management (enhanced indexation with tracking error of about 1%) makes sense and has added value in the past. Inigo showed how quantitative managers saw their share of global AUM fall from 9.5% in late 2008 to just above 5% by the end of 2013, a trend which seems to be now reversing with increased market share so far in 2014.

All presentations at the conference can be downloaded from:



[1] Romain Perchet, Raul Leote de Carvalho and Pierre Moulin. “Inter-Temporal Risk Parity: A Constant Volatility Framework for Equities and Other Asset Classes.” Working Paper (2014) http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2384583

[2] Romain Perchet, Raul Leote de Carvalho and Pierre Moulin. “Inter-temporal risk parity: an application to factor investing.” Working Paper (2014) (soon available for download from http://www.ssrn.com)

[3] Raul Leote de Carvalho, Xiao Lu and Pierre Moulin. “An integrated risk-budgeting approach for multi-strategy equity portfolios.“ Journal of Asset Management. Vol. 15, (2014), pp: 24-47. http://www.palgrave-journals.com/jam/journal/v15/n1/full/jam201411a.html

[4] CART: Classification And Regression Trees

[5] David H. Bailey, Jonathan M. Borwein, Marcos López de Prado and Qiji Jim Zhu. “Pseudo-Mathematics and Financial Charlatanism: The Effects of Backtest Overfitting on Out-of-Sample Performance.” Notices of the American Mathematical Society. Vol. 61, No. 5 (2014), pp: 458-471.

Raul Leote de Carvalho

Deputy Head of Quant Research Group

Leave a reply

Your email adress will not be published. Required fields are marked*