Screener vs Factor Model: Which One to Use

Stock screening has been one of the go-to tools for equity managers and stock pickers. Screening is indeed a powerful tool that helps you identify the right stocks and build the right investment strategy.

However, screening has its limitations. The biggest limitation is that you can miss out on quality stocks. Suppose you want to screen stocks that have a market cap greater than Rs 10,000 crore and a P/E of less than 15.

Below is the list of stocks that qualify in this screen:

There are 80 stocks in this screen. However, if I reduce my market cap criteria to Rs 9,500 crore, there are 84 stocks on the list. It is quite possible that some of these new stocks are worth considering but were filtered out due to my Rs 10,000 crore (somewhat arbitrary) criteria.

Similarly, if I increase my P/E criteria to 15.5 (from 15), I get 83 stocks. Again, it is quite possible that these 3 new stocks could be worth considering but were filtered out due to my 15 P/E (somewhat arbitrary) criteria.

The second limitation of sequential screening is that you are not in control of the relative importance that you want to assign each metric. If your filters are based on absolute comparison (e.g., P/E < 15), all metrics will get equal importance regardless of the order.

Instead, if you use percentile-based filters (e.g., P/E is in the bottom 25 percentile), metric importance reduces with the order in which they are applied. At the same time, you don’t really have much control over quantifying the importance, apart from specifying the relative order.

Using the Factor Model for stock selection avoids both these limitations.

• You don’t miss out on stocks that are at the boundary of your filter criteria
• You have full flexibility to define the relative importance of each metric

So, in the next article, we will learn how to build Factor Models. This feature is the first of its kind for Indian retail investors. So are you excited?