, , , , , , ,

In the last post, the model was generating pairs with unique relative scores that can be tracked through time. Let’s now focus on building in a baseline for those pairs.  Each pair of nearly identical pairs needs to have a baseline score of the average in the sector, so we can generate the best offsetting pairs.

This post is about the Sub-Sector Scoring Engine (SSSE) and how it will be incorporated into the model. This is the key module toward generating future trading pairs in the model that are not currently deployed.

The SSSE is reviewing over 6,000 stocks, divided into 217 sub-sectors, and updates its data daily about one hour after market close from www.finviz.com, allowing us to track changes over time.

The engine was designed and coded by my friend, Samir, at investexcel.net to generate each of the sub-sectors in U.S.-based equity markets. He was kind enough to add new features to his original Stock Screener model available here.

The SSSE gives a mean score for each of the 40+ factors we use in the model to track changes between two stocks. It generates an easy-to-use strategy for reviewing large numbers of nearly identical companies.

We can then take the comparisons and rank each of the underlying stocks, so we have a best- and worst-case score for each of them.

We then can compare and contrast the performance of these pairs, to see which have been trending away from each other at the highest rate of speed.

The screen capture above is the REIT Residential group, which has 26 equity selections in it. The model will compare and rank each of these stocks based on their average score in the sub-sector.

The obvious premise is we want to have the SSSE highlight the most divergent pairs to combine. It’s one thing for a model to have unique pairs that generate results; it’s completely another thing to have the model actively seeking to find the next best pairs to be used.

The SSSE gives us the ability to dive into the equity markets and find the most logical Long/Short stock pairs based on their ongoing results. The data is updated daily, as fundamentals & technical factors change through time.

Again, our long-term goal is moving these proof-of-concept Excel files to an industrial database that allows us to track each stock’s score against the others in its sub-sector.

We’ll be able to track historical performance and results between stocks to find the best performing pairs. If you have any suggestions on how to improve this model, please do so in comments.

Jack H Barnes Jr.

CEO & Founder

1 (800) 658-7572 Phone
1 (888) 370-7610 Fax