Quant: Man vs Machine

Related Categories

Andrew Kaplan offers his reflections as a fundamental value investor who found himself working at a ‘quant shop'

Working at a ‘quant shop', the cynic might expect me to be a proponent of all things quantitative; clearly it is in my best interest to approve and advocate. Well, I have certainly drunk from the fountain, but will it preserve my youth?

My youth was spent in New York City. I was too young and uninterested in the early 1970s to understand the apathy and recession of that era, but I was blessed with the biggest bull market in history coinciding and correlating with my first investments in the early 1980s. Fundamental analysis was a straightforward, logical method to access, understand and disseminate value. It just made sense. Reading Graham and Dodd's classic ‘Security Analysis' way before I should have, and speaking to elder statesmen wearing ‘hindsight goggles', made it clear that fundamentally derived value was the way to go.

Wasn't it obvious to all of us when, in 1983, Warner Communications was selling at 50% below its highs, that this was a content-laden company that Mr Market was offering on the cheap? I got on my bike, rode downtown to 75 Rockefeller Plaza, and elevatored my way (before post-9/11 security) to the main offices to request an annual report. On my way out, intrigued by my appearance, the man who overheard me questioning a receptionist who could not answer told me to read the documents and feel free to call back with any queries.

I never did call Steve Ross back, but I did invest. The subsequent upswing in WCI, and the rising tide of stocks over the next few decades again allowed one to think that this was clearly the way to go. I was too engaged with normal 21-year-old behaviour and too minimally invested in 1987 to care about Black Monday, and after reading Seth Klarman's Margin of Safety a few years later (‘Risk-averse value investing strategies for the thoughtful investor'), there was clearly one road to take.

Then came the bubbles of the 2000s.

The drops and beatings received in the first decade of the new millennium were a wake-up call to all. Was there a panacea to the punishment of complacency and greed? Were there better analytics to engage; or were there better analysts to employ? Clearly discipline was lacking in the canyons of Wall Street. Is it better to employ a living, breathing decision maker, or does the very fact that he is human make him susceptible to errors of emotion? Maybe there wasn't a panacea, but certainly there was a different approach. With the growth, speed and truthfulness of computers, quantitative analysis not only became more accessible, but more accurate.

So is it better? It is the classic debate between deontologists and teleologists. Either you believe that rules are rules and you stick to them, or you use a moral compass for orientation. In a true quantitative and systematic approach, the computers and algorithms spit out the results, and the trade is made. There is no room for discretion. Discretion puts you in the teleological camp. If you are adding a discretionary layer prior to trade execution, then you are filtering via human emotions, and although human emotions are not what got us into the mess of the 2000s, they certainly exacerbated its demise as the sell orders were shouted across the canyons when people saw their nest eggs crumbling.

Before I joined the quantitative shop where I have been for the last three years, I only knew fundamental analysis. I was not only concerned about how I would explain a product developed, maintained, and tweaked by eight PhDs, with countless other advanced scientific degrees - I also wondered whether I really believed it.

Let's take a step back and re-examine how the two schools differ. There are many different levels and ways to employ quantitative analysis. Quantitative analysts can cover thousands of companies in absolute real-time. They don't care if you're selling widgets or waffles, all they care about are the numbers. Various data are fed to models or robots, and depending on the algorithmic recipe, the meal is served and should be eaten no matter how appealing. What if some of the ingredients were spoiled? What if there is too much overcrowding in the kitchen? Also, were their individual recipes growing old or were they constantly upgraded, keeping up with the world's changing palate?

Fundamental analysts start with the Graham premise of taking apart and looking at the pieces of a business. The analysts may use computer programs and screen-selection to prepare their meal as well, but there are quite a few ingredients that the computer is unable to digest. For instance, did the CEO stumble in his delivery during the earnings call? Is there value in a brand? What is in the R&D pipeline?

Once the qualitatively-derived meal is presented there is yet another decision process. Should I add salt? Does it need to cool off a bit first? It is a rational approach, but there are problems. While not everyone is using the same ingredients, aren't they all using the same cookbook? Very few have the true academic and intellectual (let's call it culinary) advantages to create a unique and superb combination. And what if that talented chef was allergic to, or just didn't like, mushrooms or onions? Finally, depending on the hunger level and appetite of the global restaurant patron or waiting for the temperature to change could mean the difference between a one and three-star rating.

There have been many published papers by experts all clearly more erudite than me. They analyse the pros and cons of the two strategies, discussing how these strategies perform over the short and long hauls. In general, fundamental shops seem to produce higher returns than their counterparts over time frames of 20 years; yet quantitative funds appear to have provided similar five-year performance with more attractive risk-adjusted returns - higher Sharpe ratios and downside protection. I just attended a conference last month in New York where the pension funds on the panel unanimously agreed that they were not searching for double-digit performance - they just didn't want to wake up one day way to the left of the bell curve. Protect their downside, and they are happy - or so they said at this conference.

But what about the emotion? Emotion drives us in all we do. Some is apparent, some is hidden, but we are human. We are fallible. Therefore, we are irrational. Quants seek to bring rationality into the financial arena and strengthen a world long dominated by instinct, guesswork and fallacy. This doesn't mean that quants will ever be infallible. The scientific method is applicable to finance, but it is not as robust in the financial sphere as it is in nature - after all, there is only 10-20 years of recorded high-frequency data available for study in most financial futures markets, as opposed to centuries-worth of recordings in the natural world. It is, though, still a valid field of enquiry because there is no other truly rational way to approach the markets.

The scientific method and statistics remain the best way to forecast the behaviour of financial systems. They also distance the investor from the emotions that undermine much investing. There are at least two basic behavioural patterns that define human investors: over-reaction to bad news and under-reaction to good news. This, at least, partially explains the existence of mean-reversion and trends in the financial price data, both of which invariably lead to mistakes and losses. Ben Graham, the father of modern fundamental analysis even warned of the danger of emotion when he said that "even the intelligent investor is likely to need considerable willpower to keep from following the crowd".

In my opinion, both quantitative and fundamental strategies provide benefits and can add value to an overall portfolio - but what I have learned over my three years at a quant shop is the value of a quant.

Andrew Kaplan is director of marketing at Systematic Alpha Management, a New York-based quant managed futures specialist


Have your say

You must sign in to make a comment


Your first step in manager selection...

IPE Quest is a manager search facility that connects institutional investors and asset managers.

  • QN-2503

    Asset class: Equities.
    Asset region: Emerging Markets.
    Size: EUR 30m.
    Closing date: 2019-01-31.

  • QN-2505

    Asset class: Real Estate Core/Core-Plus Multi-sector strategy.
    Asset region: Asia-Pacific.
    Size: $ 50m.
    Closing date: 2019-01-28.

  • QN-2506

    Asset class: Currency Overlay.
    Asset region: Global.
    Size: USD 4 bn..
    Closing date: 2019-02-11.

Begin Your Search Here