InsideArbitrage Now Has a Quant Model

The “Jenga” scene from the movie The Big Short is a favorite of mine. The part of the scene that really cracks people up is where Ryan Gosling says “that’s my quant” when asked about how he could be so confident about the pitch he was making (yes, it is very racist).

Over the last several weeks, we had a lot of fun building a new quant model at InsideArbitrage and I asked several friends and investors to check out early versions of our model. Their valuable feedback and some tweaking on our end finally resulted in a model that I have a lot of confidence in and we rolled it out on our quotes page during the Thanksgiving weekend.

We decided to call the quant rating system IA Score and you can see it in action on the Lululemon Athletica (LULU) stock quote page below.

To understand why I decided to build a model, check out the article I wrote in September titled Perplexity and Multi-Factor Stock Rating Models where I mentioned:

“Earlier this week, I once again embarked on building a multi-factor stock rating system (without first peeking at that old model from 2010). Yes, we all know that there is no substitute for qualitative analysis. However I come across many new companies every single week as I look at insider transactions, new M&A activity, stock buyback announcements and spinoffs and wanted a quick filter to identify companies worth exploring.

This new rating system encapsulates my current manual process and in the coming weeks I will play with both the factors and their weighting as I run the model through dozens or hundreds of stocks.”

In a past life, I had back-tested a large number of strategies and even ran machine learning models (rudimentary models compared to what is out there now) to come up with a set of ideal strategies that outperformed their benchmark.

“If you torture the data long enough, it will confess to anything you want.”

This overfitting problem is a key issue with back tests and one of the reasons factor-based models have failed in the past.

This time around, I decided not to do any back testing to come up with this model and instead used my intuition to pick both the factors and the weights assigned to them. The other change from my old 2010 model is that I used several forward looking factors and included trends related to revenue growth and profitability. We also decided to allow investors to remove momentum or valuation factors and see how the model performed without those factors for their stocks.

For example, the company Energy Services of America (ESOA) that I discussed in my article Sixty Companies, A Two Day Conference in Dallas And Ten That Made The Cut scored 98 out of 100 in our quant model. The past year has been particularly kind to ESOA investors with the stock up 222% and it is up 1,937% over the last five years. Is the company scoring well on the model just because of the momentum in the stock? Removing momentum factors drops the score to 93 and removing valuation factors drops the score to 94.

We then ran the IA Score model against the entire universe of U.S. stocks and below is a list of 50 top stocks ranked highest by the model. Quant models can be inherently flawed and I use them as the first step in my process. So please do your own due diligence and note there may be microcap companies or penny stocks in this list. The model is not set in stone at this point and this is version 1.0.

How does your top idea rate on our model?

Disclaimer: I hold a long position in Lululemon (LULU). Please do your own due diligence before buying or selling any securities mentioned in this article. We do not warrant the completeness or accuracy of the content or data provided in this article.