A Look into the Quant World

A discussion with Savita Subramanian, Head of U.S. Quantitative Strategy for Merrill Lynch.

How did you get started in quantitative strategy?

I attended the University of California Berkeley and graduated with a degree in math and philosophy. When I finished I ended up on the buy side at Scudder Kemper in their quantitative products group. The group I was in managed money using quantitative models. It was my first experience in this area, and it wet my appetite for quantitative applications to finance. It was a perfect blend - statistical analysis and the more qualitative, behavioral aspects of finance. It dovetailed well with what I had studied in school.

At Scudder, I realized I had the math skills, but needed more finance and accounting skills. This realization coincided with the peak of the bubble in 2000. I went on to business school late that year, when the job market soured. I went to Columbia and specialized in finance and really liked the classes that dealt with capital markets, arbitrage pricing theory and other more theoretical concepts in finance. Also of interest was a course in value investing - how to pick "inexpensive" stocks. It was interesting to learn how inefficiencies in the market could be identified using quantitative and qualitative tools.

At Columbia, I did a summer internship with Merrill Lynch's quantitative strategy group. It was a great experience, largely because of the people I worked with. I believe that it's critical to find the right people to teach you.

I worked part time at Merrill while I was in school, and since then it has been an exciting and challenging ride. I started out as an associate in the quantitative strategy group and now, thanks to a firm willing to develop and promote talent from within, am the lead analyst.

Could you describe a typical day?

On a day-to-day basis, my work consists primarily of building models, which requires experience with model inputs, construction and attention to detail. The most important skills are an understanding of financial concepts coupled with quantitative skills. You work with data regularly and must be in step with interpreting and managing it.

Next, this work involves the development of ideas. We have to come up with investment ideas for our customers. In essence, we look for pricing inefficiencies in the market and unique strategies for stock investing that we can share with the firm's clients.

Finally, we talk to clients about different trading ideas and strategies. This is often the most rewarding part of my time spent, because clients' input often leads to idea generation and helps in coming up with relevant strategies. Now, as lead analyst, I also am cultivating my managerial skills, whether it comes to hiring or day-to-day management.

I still keep trying to come up with ideas, and work with our sales team to communicate equity strategies to Merrill's clients. We're also building new models, for example, how to pick stocks in specific industries that are adaptive to changing investment environments. Building a model takes up the greatest amount of time. Once a model is built, it's more a matter of plugging in the right numbers.

Having said that, we have to constantly monitor the models' continued efficacy. I generally spend a third of my day writing research for clients, a third engaged in client interaction, and a third conducting the analysis.

We publish about 10 regular research reports on a monthly basis, and then more topical reports about once every week or so to capture current market trends and points of interest. For example, we might look at why long/short investors may be having a difficult time finding opportunities in the market - well, the spread between the best and worst performing stocks is at an 18-year low!

What advice would you give to potential Quantitative Strategists?

Someone who wants to work in quantitative strategy should find a firm with good mentors and a willingness to develop and invest in them. The typical track is focusing on a quantitative discipline in college, such as engineering, mathematics or statistics. Next, work for a few years and then go back to school for a master's or Ph.d. in some area of quantitative finance. There are many Wall Street firms that are actively seeking Ph.d's in finance, economics or specially tailored mathematics in finance programs.

Are there specific resources you recommend?

Some publications I would recommend include the Financial Analyst Journal and the Journal of Portfolio Management. There are also conferences and organizations you can participate in. The SQA Society, or Society of Quantitative Analysts, holds events that are focused on topics in quantitative finance like modeling risk, portfolio construction and optimization, building factor models and the like. The CFA (Chartered Financial Analysts) Society is a great resource too, for both conferences and interesting papers. And keeping in touch with what academics are doing is useful in staying ahead of the curve.