Can I do it too? Learning Quant

My first experience of being a quant was when I worked for a hedge fund dealing in commodities years ago. Ironically, it was a matter of luck that I got the job as I didn't have any formal training in mathematics or programming back then. My background was more in the financial markets. However, the experience got me started on my multi-year learning journey in quantitative finance and I realized that people who want to embark on a similar route need to know that this is a VERY VERY diverse field.

Most quantitative finance programs in universities focus a lot on the pricing and risk management part of finance. You learn probability, statistics and advanced calculus to price financial instruments or to estimate how much risk your company is taking. Such advanced knowledge and skills are needed because companies are trading increasingly complex derivatives which are hard to value and trade. This in turn makes it harder for risk managers to estimate the (probabilistic) losses that a firm can incur on any single day.

On the other hand, there is an even more arcane part of quantitative finance, concerning the prediction of market prices and direction, which is not really taught in business schools yet. Increasingly, people are implementing statistical models to predict where the market will be heading next. These models typically draw on knowledge from a variety of fields such as artificial intelligence, computer science, statistics and signal processing. While there are available papers published on them, many people usually choose not to publish such strategies. I provide an example of such a paper at the end of this article.

To complicate matters, there is another area of quantitative finance that has very little to do with mathematics. Instead of trying to price complicated financial instruments, some engineers focus on arbitraging identical instruments, such as buying gold futures and selling gold bullion if prices between these two similar assets diverge. However, such market inefficiencies are usually short-lived. Traders usually build very fast computers to automatically spot and exploit these opportunities.

If there is one lesson that I want to share, I must say that these three areas of quantitative finance are very different and likewise require very different skill sets to be successful in them. To trade complex derivatives, you will probably need to work at an institution and have the educational credentials, such as a master's in financial engineering or a Phd to be hired in the first place. To build models to predict market direction, you will need to have read extensively to have an idea of the more successful formulas out there. In addition, you need the know how to amend and implement these models in live trading. The last aspect of quantitative finance that I mentioned will require good engineering and programming skills to reduce latency in your computer program.

These skills need not be complementary. If you are planning to be a quant, you need to consider your objectives and the competencies you want to create, i.e. know what you want and find out what you need. Unless you are thinking of working for a top institution, you do not necessarily need an advanced degree. If you are thinking of implementing your own quant trading strategy, what you need is basic knowledge in either math or programming. At the very least you need to believe that you have the intellect to complete a college degree in a non-liberal arts program. This is because both math and programming require a certain amount of logical analysis and abstract thinking. The rest of the knowledge can be acquired through hard work and patience. Over the years, I have lost count of the models that I have built and thrown away despite their academic rigor. On the other hand, I have personally met academics who started their own funds. If you want to build your own quantitative trading model, you will need to be prepared for the long haul. After thinking for some time, I realized that the easiest way to get started in trading is to match your strengths to your trading strategy.

These are some books (and links) from Amazon.com that I recommend if you are considering going "quant". I have read all of them before and I find them easy to comprehend for most people. Emanuel Derman is one of the first few quants on wall street and his book My Life as a Quant: Reflections on Physics and Finance will give you an idea of what quants do. Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets by Nassim Taleb will teach you how to think like a quant, with some basic statistics. Neural Networks for Financial Forecasting (Wiley Trading) is an old book. However, it makes a good starting point for some one to get acquainted to building trading models.

The sample paper by Neely that I promised earlier.

Like what you have just read? Digg it or Tip'd it.

The objective of Finance4Traders is to help traders get started by bringing them unbiased research and ideas. Since late 2005, I have been developing trading strategies on a personal basis. Not all of these models are suitable for me, but other investors or traders might find them useful. After all, people have different investment/trading goals and habits. Thus, Finance4Traders becomes a convenient platform to disseminate my work...(Read more about Finance4Traders)