Trending Assets
Top investors this month
Trending Assets
Top investors this month
SLT Core Portfolio: Quantitative Screener
To source new investment ideas in line with our investment philosophy, we have internally developed a quantitative screener. We built it using Python and data from the free Financial Modeling Prep API. It follows the below steps:

  • Quality-Growth Score
  • ESG Score
  • Earnings Surprise Score
  • Final Score

The code we developed for this screener is freely accessible and open-source. The full code can be found on the following GitHub repository: https://github.com/SLTResearch/slt_core

Quality-Growth Score

The initial step of the screener is the collection of the following quality and growth metrics:
Operating Profit Margin, Net Profit Margin, ROA, ROE, ROCE, Current Ratio, Debt/Equity Ratio, Debt/Assets Ratio, FCF Yield, 5Y Revenue Growth, 5Y CFO Growth, 5Y EPS Growth.

The score calculated based on these metrics is the main component of the final score. We are using a simple scoring method, where a sub-score between 1 and 4 is calculated for each metric. A sub-score of 1 for a specific ratio means that the ratio is in the worst quartile. At the other end of the range, a sub-score of 4 indicates that the ratio is in the best quartile. Each sub-score for a specific company is then added together to calculate the overall quality growth score of the company.

ESG Score

ESG factors are integrated immediately during the quantitative selection process phase, as we believe that companies with good ESG scores are more likely to have a qualitative and sustainable business model over time and throughout the economic cycle. We also believe that as providers of capital, we all have a responsibility to allocate capital to companies that take environmental, social inequality and/or governance issues seriously.

This second step is quite similar to the first one. This step consists of collecting the Environmental, Social and Governance Scores. The scoring method is also similar where points are awarded based on the position of the score relative to the group. If any of the scores are in the worst half, a sub-score of 0 is assigned (i.e. no additional points for a below-average policy/ESG score). If the score is in the better half, a positive score of 1 is awarded. The sum of all sub-scores (between 0 and 3) for a specific company is then calculated.

Earnings Surprise Score

We also believe investors are more likely to reward companies that not only surprise positively on earnings but also demonstrate the potential for consistent performance with a relatively low standard deviation of earnings surprises. This is the reason why the third step of the screener aims to compute a score based on the average of the last 5Y EPS beat/miss as well as the standard deviation of these last 5Y EPS beat/miss. Higher score will be given to companies consistently beating consensus estimates within a reasonable standard deviation.

Final Score

The last step is a simple addition of the three previously calculated scores. The following link is the screener output showing companies with both a Quality-Growth score and a final score above 40. Companies are sorted from highest to lowest final score: https://github.com/SLTResearch/slt_core/blob/main/sltcore_screener_results.xlsx
GitHub
slt_core/sltcore_screener_results.xlsx at main · SLTResearch/slt_core
SLT Core Portfolio Code Repository. Contribute to SLTResearch/slt_core development by creating an account on GitHub.

Related
Already have an account?