Feature

Why Quant Is Still Not Sexy

Quantitative trading that thrives on algorithmic models still has few takers in India

“The world clings to its old mental picture of the stock market because it’s comforting; because it’s so hard to draw a picture of what has replaced it; and because the few people able to draw it for you have no interest in doing so.” That’s the introduction by Michael Lewis in his book Flash Boys on the advent of high frequency algo trading in the US markets. He further mentions that the US stock market is now a class system, rooted in speed, of haves and have-nots. “The haves paid for nanoseconds; the have-nots had no idea that a nanosecond had value.”

Despite concerns of how sophisticated machines are ruling the roost, the US regulators have restrained from putting curbs on HFTs, which are advanced and sophisticated algos that can execute trades at lightning speed sans human intervention. But closer home, Sebi recently floated a discussion paper titled ‘Strengthening of the Regulatory framework for Algorithmic Trading & Co-location’ which envisages putting certain restrictions to limit the technological edge of high-frequency trading. Among others, the proposed restrictions include speed bumps, frequent batch auctions every 100 seconds instead of ‘continuous matching’ system, minimum resting time between orders. Resting time is time between an order is received and the order is allowed to be modified or cancelled.

Though some algo traders in India are concerned about the regulatory intervention, those practicing quantitative trading, which require human intervention, are not perturbed as they don’t necessarily have the same need for speed that the HFTs have. Quant traders run multiple strategies on a single or a wide array of asset classes to make sure their portfolios generate decent returns. And for that, they don’t need to hit orders before everybody else or get a fluid stream of tick-by-tick data flowing on their screens. They can make do with a delay of a few seconds. They are typically medium frequency traders or long-frequency traders and unlike their HFT counterparts for whom seconds can be worth gold, they typically hold positions for a few hours to a few days (MFT) or a few weeks (LFT).

Rishi Kohli, managing director at ProAlpha Advisory, belongs to this breed of traders. An IIT and an IIM post-graduate, Kohli manages Rs.1,000 crore in assets through his quantitative models. Quant strategies are typically divided into two classes: one based on momentum and the other on fundamental factors. “As we are doing fewer trades during the day, we prefer human intervention to offset chances of any algo error. While we don’t trade like HFTs, each of our trade is chunkier in size that is 20%-30% of the fund size,” mentions Kohli. He has designed 17 different strategies to get best returns on Nifty futures.

For now, Kohli's strategies are largely momentum-based. “Our algos go short when markets are trending lower and long when markets are trending higher. We lose money when the market is range-bound as there are many whipsaws during this period,” says Kohli, who claims his portfolio did well in FY15 when market was up and also in FY16 when the market was down.  Kohli is now launching a quant fund based on fundamental strategy. “So far, our strategies were just price-based quant. Fundamental factor quant is very big in the US and globally. AQR which used to be a big hedge fund at one point of time is now an asset management company with large AUMs based on fundamental models,” opines Kohli.

It’s fundamental

The fundamental factor quant model essentially involves taking fundamental ratios and factors as inputs and then creating a quantitative process on that. Kohli says foreign investors wanting to invest in India’s growth story are looking for this type of a product.

The offering would also have a dynamic Nifty and currency hedge instruments that will kick in when either the currency depreciates or the markets are under pressure. However, these hedges will not cap upsides as it does during stable to positive market or currency environment. With soft commitments of $5 million, the launch is slated for October 1.

For CapVeda’s Kalpesh Kinariwala, who is now based out of Dubai and currently manages assets worth $45 million for some family offices, his interest in quantitative investing piqued during 2008 crisis when he started learning more about hedge funds who were the “major contributors to the collapse”. “When the crisis broke out, I was intrigued by the financial muscle the hedge fund guys had built and started studying their models. And what came out loud was that they followed emotion-less trading and built a structure around it. This motivated me as a fund manager to build a structure that scientifically approaches the market without any emotions. I developed a quant strategy, hired a statistician to write the code for me and the algo has been running on its own since then,” reveals Kinariwala.

CapVeda runs strategies across several asset classes – fixed income, equity futures, structured products based on commodities and trades over 12 exchanges all over the world. “We try to extract volatility out of markets. Our strategy is both long-short depending upon the signals generated by the algos. Our markets are a function of liquidity and are no more driven by any fundamental co-relations. And these bouts of liquidity give opportunity to extract volatility and capture good returns,” he adds.

Kinariwala claims his strategy managed to capture 80% of entire PSU bank fall and 60% of the PSU bank retracement. “We were able to capture more than 70% of the recent fall of Tata Motors and more than 80% of the rise of Tata Motors. In the Brexit situation, we did well because the fund went short before that. In FY13, Yes Bank gave returns of 15%, but our strategy generated 73% on Yes Bank as the stock fell from Rs.500 to Rs.250 but doubled again to Rs.500." In India, CapVeda’s quant fund trades in stock futures comprising 53 stocks with the net allocation to each stock future confined to less than 2%. “Even when there is no volatility we still do fine. For instance, 2010 was a year when there was little volatility but we managed 19% return.”

Quant Capital is another firm that has been doing a lot of work in quantitative-based research and advisory. Sandeep Tandon, MD and CEO at Quant Capital, says his firm has added another layer over its quantitative models, that is, identifying behaviour and sentiment of market participants. “We measure the long and short positions in the market to recognise the sentiment and how they are changing. For instance, Bharat Forge recently came up with a bad set of numbers. Lot of people thought it should fall, but it didn’t fall and the stock was up 12%. People used this opportunity to cover their shorts.”

Quant Capital has designed several indicators to quantify market activity and predict market movements such as quant euphoria indicator, quant bearish bets indicator, quant buying intensity indicator, and quant selling intensity indicator to name a few.

Long way to go

Despite what their proponents claim, quant still appears to be an exotic affair to the broking fraternity in India. Referring to the Sebi paper on algo, Alok Churiwala, vice-chairman of BSE Brokers Forum, says: “There are two sets of people: one, who are using this facility and a large majority who are not. Some don’t even know what this animal is. They believe that there is neither a level playing field, nor any concrete economic value-add to a lay investor.” While lack of awareness is an issue, Tandon sees a major stumbling block for quant to grow in the country. “To make quant models, you require humungous amount of historical data, going as back as 100 years and 50 years. India lacks historical data.”

He adds that it will take a while before quantitative- based approach becomes popular in the country given that institutions and HNIs are taking their time warming up to the concept. “There are very few takers for quant-based strategies in India. Beyond Bangalore, there are no takers even in traditional market savvy cities such as Ahmedabad, Calcutta or Chennai.” Not surprising that the share of algo as percentage of the combined traded turnover of both the NSE and the BSE is still limited to 4%. And if the BSE Stock Broker Forum’s reaction is anything go by, then the wait looks generational.