Monday, March 25, 2013

Understanding the Mechanics of Quantitative Trading


MetaStock SPRS Series - Week 111 - TechniTrader® Stock Discussion for MetaStock Users - Understanding the Mechanics of Quantitative Trading - March 25, 2013
By: Martha Stokes C.M.T.
One of the areas of professional trading that is least understood is how quantitative analysis actually works. Many retail traders assume wrongly that quantitative analysis is a form of stock chart analysis. It is not. One book that is circulating around the Quants is "Antifragile" by Nassim Taleb. I do not agree with all of his summations and theories, but what I find fascinating is how Quants are reacting to the book. By comparing his work to quantitative analysis, we can begin to dissect how the institutions derive and use quantitative analysis for their buying and selling on any financial market—stocks, bonds, futures, forex, index, ETFs, whatever. This is the key to figuring out what goes on in the HFT and Dark Pool back rooms.

The argument stems from the use of either a professional style momentum strategy, or choosing a mean reversal strategy which is what retail traders call “SAR” strategies. Please remember that your strategies are very simple and basic compared to Quant algorithms that trigger billions of orders in a very complex set of instructions. Often this code is many pages long with mathematical equations. The basis for all Quant work is mathematics, not technical analysis. What Quants are debating is whether a momentum strategy often used by HFTs or a mean reversal, actually lowers the risk when a "Black Swan" occurs. A Black Swan is an event that is totally unexpected and sends a shock wave through the market. This is something the professional side is hoping to minimize in terms of risk, something that they have not considered as much in the past few years, often with irreversible consequences. What the Quants are trying to determine is whether their formulas and strategies are fragile or "anti-fragile" based on the risk of a Black Swan. The problem is that if the true standard deviation is higher than estimated by a mere 5%, the probability of a catastrophic event will be increased by 5 times the original estimate. Their conclusions were fascinating and exposed the underlying problems of purely mathematical trading formulations.

The Quants are using a formula theory based on a standard bell curve, with 3 standard deviations as the mathematical premise. Recent data was tested to evaluate a negative left tail movement past 3 standard deviations to see the impact on a portfolio or trading. At 3 standard deviations beyond the norm, the losses would be catastrophic in some instances. Quants are currently of the opinion, being mathematically oriented, that Black Swans are utterly unpredictable events. I don’t agree that the markets are completely blindsided by Black Swan events. In every instance that a market-induced event turned into a Black Swan there was ample warning something was wrong, but the reality was that it is not the mathematical equations that failed but the human factor that is still present even with automated orders. Greed is the catalyst of every Black Swan market-induced event. True, non-market generated Black Swans are extremely rare events, such as the Tsunami of Japan. Most market driven events have ample warning that most traders ignore. As an example, it was blatantly obvious in 2007 that a recession of significant magnitude was coming. Every economic indicator pointed to a recession. People in the investment banks were warning it was way over the top, that the real estate market bubble was extreme, and that banks were over-leveraged. The extent of the stupidity was not known to anyone but insiders, but there were warnings.

At any rate, the goal for this year’s strategies for Quantitative Analysis is to minimize the downside risk of a Black Swan. The calculations are based on the left tail negative movement past 3 standard deviations, which would cause a devastating loss similar to what JPM experienced a while ago. Mean reversal strategies are often employed during trading range or range bound markets; however left tail risk rises significantly while profits are capped sharply. This is similar to the straddles and strangles of option trading with similar results. Although the mean reversal would be construed to be the safer strategy, market conditions in the past 3-4 months have proven that the momentum Quant strategies outperformed significantly with less risk. Therefore the argument is that the momentum strategies are less fragile and less prone to left tail risk factors. Applying this to HFT strategies for the current market provides a fascinating result. What the results showed was that although HFTs do not apply margin leveraging due to their speed of execution, HFTs do not benefit from the right tail profits, meaning they don’t receive the benefits of riding the run as a swing trader would. The HFT has lower risk because of their brief holding periods, which provide for cumulative profits quickly with minimal downside risk. This is how the HFT strategy which has a 51/49 ratio of profits to losing trades, is able to make tiny percentage penny profits on millions of dollars in trades. The results proved a couple of things:
  1. Using a mean reversal did not lower risk but rather increased risk and also cut profitability even further. This is never a good strategy.
  2. Momentum strategies offered lower risk as long as both sides of the market were traded and stop losses were used religiously. This is another reason why you as a retail trader, should be extremely proficient trading both the long trade position and the sell short trade position.
Trade wisely,

Martha Stokes, C.M.T.
Member of Market Technicians Association
Master Rated Technical Analyst: Decisions Unlimited, Inc.
Instructor and Developer of TechniTrader® Stock Market Courses
http://technitrader.com
MetaStock Partner
©2013 Decisions Unlimited, Inc.

Disclaimer: All statements, whether expressed verbally or in writing are the opinions of TechniTrader, its instructors and or employees, and are not to be construed as anything more than an opinion. Student/subscribers are responsible for making their own choices and decisions regarding all purchases or sales of stocks or issues. At no time is any stock or issue on any list written or sent to a student/subscriber by TechniTrader and its employees to be construed as a recommendation to buy or sell any stock or issue. TechniTrader is not a broker or an investment advisor it is strictly an educational service.

No comments: