Finding Alphas: A Quantitative Approach to Building Trading Strategies by WorldQuant

It is said hedge funds always are reluctant to disclose their secrete sauce, well not in the case of WorldQuant, the CEO of this renowned shop is so wiling to share and propagate his insight, that he published a book about his alpha strategies.

Definition of alpha will be skipped, I would just cite their categorization of alphas, based on type of instruments, it could be from equities, from bond, currencies, and commodities, based on time horizon, it could be

Intraday alphas: re-balanced during trading hours of the day
Daily alphas: re-balanced every day
Weekly/monthly alphas, re-balanced every week/month

Developing alpha is either by sourcing public information or building specific models to process the information, searching signals/patterns, typical sources are as follows. Alpha can also be derived from optimization via risk factor models, relationship models, and micro structure models.

Measurement of the alpha robustness are as follows:

high in-sample information ratio(IR)
god out of sample IR
works well across the trading universe
less fitting
intuitive/interesting/simple idea
works in multiple regions
small draw downs
short draw down periods

Chapter 11 Alpha and Risk Factors described “a few well-studied market “anomalies” and make the point that some alphas evolve to become “hedge fund betas” or risk factors”. need to check how much is left in an alpha after neutralizing risk factors. In our practice, a good alpha tends to yield a higher Sharpe ratio after risk factor neutralization, even though per-dollar return might be reduced in some cases.

He listed out the famous market factor (from Markowitz’s 1952 mean variance and returns, CAPM in 1960s and then to size, value per Fama French, until in 1992 momentum factor is added, even with flaws such as liquidity, publicity, these well-studied risk factors are popularly implemented across the quant investment industry, and thus has caused quant crisis in 2007 when majority of the trading models adopt these factors. Hence, worldquant quants state “In our practice, a good alpha tends to yield a higher Sharpe ratio after risk factor neutralization, even though per-dollar return might be reduced in some cases.”

WebSim seems to be a good tool to try on…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.