Mimicking the work of human analysts, machine-learning algorithms allow us to rate more than 50,000 firms worldwide.
With about 120 equity and credit analysts, Morningstar has one of the largest independent equity research teams in the world. Morningstar analysts cover approximately 1,700 equities, using a consistent, proprietary methodology that focuses on fundamental analysis, competitive advantage assessment, and intrinsic value estimation. The culmination of our analysts’ work is the Morningstar Rating for Stocks.
To complement our analysts’ work and extend the coverage universe, Morningstar recently introduced the Morningstar Quantitative Equity Rating—a forward-looking measure that is generated by a machine-learning statistical model that attempts to produce ratings and statistics that would have been produced by Morningstar analysts. The quantitative rating combines these components into a single measure.
By being able to quantitatively rate stocks, Morningstar is able to extend a useful tool to thousands of securities around the world that might otherwise not be covered. The quantitative ratings are now available for more than 50,000 companies in 86 countries that trade on 64 exchanges.
A forward-looking quantitative assessment, rooted in our analyst process, is far more useful than a data page containing historical financial numbers. The quantitative ratings allow investors to obtain a much greater breadth of the independent perspective they know and trust from Morningstar. The ratings can be expressed in absolute terms or relative to the full investment universe, each country, and each sector, which helps investors evaluate the most attractively priced stocks in each category. The ratings provide additional benefits, including the ability to analyze portfolios by aggregating data and providing daily history to track changes over time.
There are several underlying components to Morningstar’s equity rating. These components— Quantitative Fair Value Estimate, Quantitative Valuation, Quantitative Economic Moat, and Quantitative Uncertainty—set out to replicate the projections of our analyst team as accurately as possible on a variety of decisions. After estimation, they are combined into our overall recommendation, the Morningstar Quantitative Equity Rating.
The quantitative rating is our summary rating and meant to be Morningstar’s best guess at the future expected return of those stocks. But how well does the quantitative rating forecast future performance? In this study, we evaluated the effectiveness of the quantitative rating at predicting future stock returns. We found that Morningstar’s quantitative ratings are able to sort stocks according to their future risk-adjusted returns monotonically–higher-rated stocks have better future performance.
Methodology and Explanation of Process
The Morningstar Quantitative Equity Rating is assigned based on the combinations of the Quantitative Fair Value Estimate (sometimes referred to as fair value) of the company dictated by our model, the current market price, and the margin of safety determined by the Quantitative Uncertainty rating. We will briefly discuss the inputs that determine the quantitative rating for a stock. Quantitative Valuation for Stocks
Morningstar’s Quantitative Valuation scores are designed to algorithmically replicate as closely as possible the valuations assigned by our team of equity analysts. That is, we would hope that an analyst would arrive at the same rating for a stock as that given by the quantitative process. With this goal in mind, the quantitative model is empirically driven and based on the valuations assigned by our analysts.
In broad terms, the essence of our quantitative model is to pinpoint the characteristics our analysts use to differentiate between overvalued and undervalued stocks. This is done with a machine-learning algorithm called random forest to fit a relationship between our fundamental and market inputs and the variable we are trying to predict (the analyst-assigned valuation of a given stock).