The lightspeed re-boot that is quant investing
SB VEDA <MUMBAI / CALCUTTA>
“The secret to being successful from a trading perspective is to have an indefatigable and an undying and unquenchable thirst for information and knowledge.”
-Paul Tudor Jones, CEO Tudor Investment Corporation, Philanthropist and Conservationalist
For around eighty years, now, economists and finance experts have been applying numerical methods to try and ‘beat the market.’ The challenge, of course, was that markets were supposed to be unbeatable. According to the Efficient Market Hypothesis (EMH) markets efficiently reflect all publicly available information, thereby eliminating any advantage for a single investor over any others. So, finding an ‘anomaly’ in this supposedly efficient market became the game to realize returns that belied EMH. Sometime within this nascent experimental period, when quantitative methods were being developed to find anomalies, quant or quantitative investing grew as a branch of financial economics.
What do Quants or Quantitative Analysts do? In a nutshell, they are developers who devisee complex mathematical models to detect anomalies, upon which transactions can be based to realize abnormal profits. They automate impute assumptions, based on market factors and choices into the model, now called an algorithm and automate this.
Anomalies don’t last long in markets, and so identifying them using mathematical models and trading on them must be done quickly – in today’s markets, lightning fast. I can attest to how difficult and time-consuming this was before the software made it easy to harness the processing power of computers.
Indeed, computers can do, today, what. Quant experts could in do in the analog age at an incredibly accelerated pace. The quant investor is still needed to devise the algorithm that finds the anomaly. In doing so, the quant establishes its theoretical basis as a result of analyzing the market, and employing econometric and other applied mathematical relationships (which really form the basis of these models) to the market info. – but computing power has been (dare I say it) the game-changer in quant investing. The Quant, even in this digital age, is only as good as the developer behind it – as with most things digital. This may change as Artificial Intelligence capability evolves in its application to quant investing.
Let me give you an old-world example of quant investing – its benefits and pitfalls – from my own experience: This was before algorithms were widely applied to a portfolio of stocks. As a grad student, my roommate and I had been studying the then hot telecom market. These were the heady days of Cisco, Nortel, Luscent and Motorola – a time when the United States produced a little over one-third of the world’s telecommunications equipment to service a market that was relatively contained compared with Asian giants, China and India. To put this in historical perspective – Verizon hadn’t even been born yet with its legacy parents, Bell Atlantic and GTE, still slogging it out in the telecom war that ensued after deregulation of Ma Bell.
With competition in telecom ramping up, my roommate and I were analyzing the telecom equipment market and found a company that had previously been successful at digitizing the radio-based taxi dispatch service throughout much of Canada. They were also into modems and network communication. Recognizing that, as modems and the building of Local Area Networks (LANs) were becoming commonplace and the infrastructure for the internet was being created, they began to focus on low-cost telecom technology for the wireless market, to be marketed to the big telecom companies, which were battling it out for supremacy in the communications sector.
With a healthy balance sheet, and having invested in the requisite R & D, many customers seemed interested. We analyzed their financials over a few years as well as how they were spending their returns. We also analyzed the market into which they wished to enter, and concluded that they had a price competitive advantage.
My roommate and I invested the princely sum of $3,000 (from our OSAP loans), which at an economical price of a dollar a share, got us 3,000 shares. Within a month as contracts were signed, the stock price rose to $2 per share. We’d expected as much with the new business expanding their market and revenues.
Then something strange happened: having automated stock movements using alerts, our computers were flashing all kinds of colours. The stock had jumped up another dollar in a few hours. The following day, it opened another dollar higher and began a meteoric rise. Recognizing that something beyond our quant analysis was going on, we sold at around $4/share, making ourselves an abnormal 400% profit. Had we ridden the wave, we could have made a 2000% return as the stock peaked at $21/share. Imagine that!
That was before the fall… After reaching $21/share, he stock took a near vertical tumble, reaching a terminal velocity of close to what we bought It for. The stock was trading at its ‘normal’ price of between a dollar fifty and a dollar sixty. We actually only intended to make around a 50-60% return – that would have vindicated our analysis, and helped us with rent an expenses for a little while.
What had happened to influence this stock’s behaviour: talks of the company being bought by AT&T were bouncing around the Street. And, speculators were buying in bulk, driving up the price. When the talks fell through as one of the company’s large competitors, Cisco negotiated a play with Ma Bell, the price tumbled as speculators rushed to realize their profits. But selling as a stock is falling can be precarious.
It just so happened that my roommate (though we didn’t know it then) was something of a gambling addict. And, he was hankering to let it ride! Aces high…sixes up! (Add to that, whatever other gambling jargon I’m just making up off the top of my head.) He argued that our quant was too conservative, and that investors had beaten us to the punch and started pouring their money into the company as a result. I maintained, firmly, that the analysis was sound, and that we had to be ‘rules-based’ and sell at $2. That the market had already surpassed our sell mark meant we had to sell right away as a correction was forthcoming. And, corrections can be over-correcting, eating into expected profits, sometimes even giving rise to losses. Well, fortunately, I managed the e-broker and sold us out.
This, of course, caused quite a furor between us as the stock continued to rise. But my roommate would have hung on as the stock peaked and then plummeted, trying hang on for a recovery while the stock continued to sink like a stone. He might have ultimately been able to sell at our sell mark. But, as it happened, my quick action caused us to to better. Sure… we didn’t turn $3000 into $63,000 – but quants aren’t about winning the lottery. It’s about making sound rules-based decisions on the data at hand – not hitching on to tips or riding rumours.
In the end, we made an abnormal profit in part from our analysis and accidentally from the rumour about which we were oblivious at the time. We bought on the rise because I knew that what was going on was not ‘rules based,’ which is the Bible for Quant investors. Something else was driving up the price beyond what was rational and we wanted no part in the gambling notwithstanding my roommate’s dopamine receptors having caught on fire!
Would an algorithm have helped us? Probably not – but, absent my discipline, it certainly would have forced us to sell before the price went below our desired optimum sell-point.
Information is the key in either case – quant investing or the murmurs of the Street: As Paul Tudor Jones once said – “The secret to being successful from a trading perspective is to have an indefatigable and an undying and unquenchable thirst for information and knowledge.”.
We were using good old Warren Buffet’s value-investing principles, which we learned in Finance 101. We saw a stock that we felt was undervalued and poised for a revenue surge, and we bought low. We sold high – higher than we thought we should. And, then we moved on.
But today’s Quants go further. They engage in, among other methods, factor investing. This is simply, investing based on utilizing multiple factors, including macroeconomic as well as fundamental and statistical characteristics to analyze and explain asset prices and build an investment strategy. Factors include such elements as: macroeconomic factors such as GDP, inflation, market capitalization and the employment rate as well as growth vs. value; credit rating; and stock price volatility – to name some. Factoring characteristics in the Algo is as much art as science and depends on the assumptions and judgment of the Quant.
Another is replication. Quants can design Algos that mimic the trading patterns of successful traders. This is a perilous strategy because, at the end of the day, a cubic zirconium however much it glitters is no diamond!
Actually, the number and types of strategies is virtually unlimited and based on the intelligence and creativity of the Quant.
In our case, an alga-based approach would have analyzed the whole market, taking a wider angle view in devising a strategy – and likely recommended incorporating a portfolio approach into our very narrow analysis. It would have engaged in a more thorough assessment of competitors in the market and factored in their respective positions in the market. Maybe a Quant would have priced the stock at $1.25-1.30, lower than our projected price. The company was an upstart, and we were influenced, too by management. Sometimes, it’s good to use judgment to bend the rules in rules-based investment
Quant investing is making a comeback because, a retreat occurred when markets were booming and throwing a dartboard at the S&P were yielding good returns – i.e., there were easier paths – and hedge funds, the market makers found themselves making a killing just by buying big. Big Tech, opened their coffers and began to gobble up stocks in a way that dramatically influenced benchmarks and indices. Investors, looking to make a quick buck, mimicked the big boys, riding the waves of Blackrock and others. Well, the post-pandemic reversal has taken place and Blackrock is laying off by the thousands, showing its first bad quarter in a long time.
So, the lonely Quant is now the miner of the anomaly – that sometimes small difference that is uncovered in analysis from publicly available information that predicts the movement of a stock.
Computers make it happen at Millennium Falcon light speed – but we still need the Quant to say, “Chewie, punch it,” before the thing rockets into hyperspace. Time to reboot our analysis of companies and markets. The Quant is back in a big way!