Alpha AI. Systematic quant research for every holding.
Alpha AI scores thousands of stocks every night across five factor families, freezes every metric point-in-time and produces a deterministic, traceable score – the same engine that powers your rankings.
Research, not investment advice
Research at an institutional level
No black-box gut feeling. Alpha AI makes every score traceable – from metric to calibration.
Five factor families
Momentum, value, quality, growth and analyst consensus are blended into a weighted composite – each dimension normalised to −1…+1 so no single metric dominates.
Point-in-time archive
Every night, each security's raw feature vector is frozen as of that date. Backtests stay free of look-ahead bias – the foundation for honest signal research.
Deterministic & reproducible
Live ranking and the nightly precompute share the exact same scoring core. The same input always yields the same score – no noise, no drift.
IC-calibrated weights
Factor weights are principled defaults, sharpened against the information coefficient and validated across multiple market cycles and decades of market data – not fitted to a single year.
The five factor families
Every holding is scored along these dimensions – transparent and individually traceable.
Momentum
Trend strength from technical analysis: moving averages, RSI, MACD, Bollinger and cross signals.
Value
Valuation via forward and trailing P/E, price-to-book and dividend yield – judged relative to sector and history.
Quality
Profitability and balance-sheet strength: return on equity, net margin, leverage, free cash flow and liquidity.
Growth
Revenue and earnings growth year over year – momentum, not a snapshot.
Analyst consensus
Sell-side consensus ratings and price targets, with an upside veto: unrealistic targets cap the rating.
More than a score: valuation discipline & signal depth
A high score alone isn't enough. Alpha AI knows the traps that simple metrics miss.
Cyclical peak trap
A low P/E at the earnings peak of a cyclical isn't a bargain – it's a trap. Alpha AI recognises peak earnings and reads valuation through the cycle.
Bubble warning
A premium valuation without cash flow is a warning sign. When a name is expensive and burning cash at the same time, that gets named – not sold as growth.
Cash-flow quality & balance-sheet risk
Free cash flow, leverage, share dilution and other balance-sheet signals feed the quality assessment – beyond the plain P/E.
Flows & earnings drift
Insider and institutional buying, short interest and post-earnings-surprise drift add signals for when a thesis is working.
From raw data to a reproducible score
Four steps, every night, across the entire stock universe.
Score the universe
Alpha AI scores thousands of single stocks nightly across STOXX 600, S&P 500, Nikkei 225 and MSCI Emerging Markets.
Archive point-in-time
Each security's full feature vector is stored as of that date – the data basis for backtests without look-ahead bias.
Rank cross-sectionally
Factors are normalised, blended into the composite and translated into a 1–5 star conviction and a clear verdict – including risk flags.
Calibrate against forward returns
As the archive matures, factor weights are sharpened using the information coefficient – research that improves over time.
See Alpha AI in your portfolio
Ask your co-pilot about your holdings – backed by systematic quant research.
Alpha AI is a research and analysis feature and does not constitute investment advice. Factor weights are undergoing ongoing IC calibration; scores are not a promise of future performance. Simulated or past results are not a reliable indicator of the future. You make your own investment decisions.