Value Investing Analytics

Methodology

The live ranking uses a hand-set factor ensemble as the published score, with two experimental comparison models shown beside it: a Revenue/Shrink/Assets combo and a robust Huber regression model. We tested the factor set on a broad in-house universe from 2008 through the latest date with a completed forward-return label, using forward total return rank as the outcome and normalized cross-sectional factor ranks as the inputs. The goal is simple: keep the live score readable and stable, while still learning from the model research.

Current Live Score

Gross Profitability to Total Assets
12-Month Momentum
Year-over-Year Share Shrink
3-Year Revenue CAGR
Year-over-Year Revenue Growth

These are the same five factors used in the earlier Custom Research Ensemble. The published score is still a transparent weighted mix, but momentum now carries less weight than before.

The published Final Score stays on a clean 0-1 scale because it is built from percentile-ranked factor inputs. The experimental model scores also stay on a 0-1 scale by percentile-ranking the combo score or model predictions across the current universe.

We still show P/E Ratio and Earnings Yield in the table because they remain useful diagnostics. They are not part of the published Final Score.

Why keep an experimental model view

We first used research to identify a strong five-factor set. Gross profitability, share shrink, momentum, and revenue growth repeatedly showed up in the better combinations, so they became the core ingredient list. After that, we tested whether smarter combination rules could do a better job than fixed weights while still using exactly the same inputs.

Recent walk-forward tests showed that the best live comparison candidates were not the old LightGBM column. The Revenue/Shrink/Assets combo had the strongest top-decile compounding in the standard reports, while the Huber All11 model was the strongest robust ML-style candidate. Both are shown for comparison and are not used to set the published rank.

Backtest setup

Window: Begins in 2008 and ends at the latest date with an available forward-return label for the selected horizon.
Target: next 36-month total return, including splits and dividend reinvestment via adjusted prices.
Ranking framework: factor ranks and forward return ranks were both normalized cross-sectionally by observation date.
Validation lens: mean rank IC, top 20% versus bottom 20% spreads, top 10% versus bottom 10% spreads, and hit rates.
Ensemble validation: model methods were evaluated with an expanding walk-forward setup, so each scored date was predicted using only earlier dates.

Mean rank IC by method — The newer method reports compare static combo candidates, dynamic factor weighting, and robust ML-style models against the published score.

Top 20 percent minus bottom 20 percent spread by method — The top 20% versus bottom 20% spread chart shows how much each method separated the strongest names from the weakest names on average.

Top 10 percent minus bottom 10 percent spread by method — The decile spread view makes the sharper separation visible: the best nonlinear methods widened the gap substantially more than the hand-weighted score.

Method correlation heatmap — The method correlation heatmap shows which ensemble methods are producing genuinely different rankings versus mostly rephrasing the same ordering.

$10k top-decile quarterly rebalanced growth by method — The top-decile growth chart shows what a quarterly rebalanced portfolio of each method's highest-ranked names would have done.

The decile chart now reads from lowest-rated stocks on the left to highest-rated stocks on the right.