Quantitative Equity Research vs. "Normal" Equity Research

There are several approaches to quantitative equity research some focus on individual security returns others may focus on the performance of the portfolio as a whole. If you're evaluating an individual security you may find a basket of similar securities and then find out how certain signals (financial ratios, cash flow patterns, insider buying, GDP growth numbers) are correlated to the return of a stock's price. Momentum strategies are also popular, they state that if the return of a stock or an index has gone up over these last couple of months then it will probably keep increasing in the future. Conversely, if the stock price has been trading lower and lower it will continue. If you're managing a whole portfolio clients may also want to implement portoflio construction techniques that minimize quantifiable risk and max returns.

These quant positions require the use of Matlab, R, SAS, STATA, C and at least a masters degree in something heavily quant (math, cs, engineering, physics) if you're going to be doing anything remotely interesting.

"Normal" Equity research involves looking at financial and qualitative information that you can gleam from the company, either through their SEC filings like 10-k or 10-q documents. Looking at their financial histories, looking at the strength of financial metrics in balance sheet, cash flow and income statements to see if their assets are making the requisite earnings to justify a price. Other things that they might do is look at the quality of their managers, the quality of their supplier, distribution and buyer networks, the competition in their industry, competitive advantages in their product, how would they do if event A occured as opposed to event B etc. Some investment companies build DCFmodels and others may use comps to determine whether a company is priced correctly according to the market and will act on that.

I've never heard of pure normal equity research using R, Matlab or any other kind of advanced statistical analysis packages to analyze data. Although they can definitely use this software to augment their current efforts.