ML Market Map — Hong Kong Equity Clusters for 2026-08-31

A daily unsupervised machine-learning read of the Hong Kong Equity market: 2,448 stocks grouped into 9 consensus clusters (KMeans + Gaussian-mixture + hierarchical, over robust-scaled PCA features) for 2026-08-31. Descriptive, not predictive — there is no buy or sell signal. Research, not investment advice.

Markets: US Equity · China A-Share · Japan Equity · Hong Kong Equity · Taiwan Equity · Korea Equity · UK Equity · India Equity

Today's clusters

broad market · slight falling earnings — 624 names; mostly Industrials; drivers: ni growth (−0.33), pa mae 21d (−0.28), pa mfe 52w (+0.27)
large cap · rising earnings · high margin — 572 names; mostly Industrials; drivers: r2 252 d (+0.81), log market cap (+0.70), ni growth (+0.59)
high turnover · attention spent · heavily traded — 250 names; mostly Healthcare; drivers: turnover to mcap (+3.57), pa sir recovered (+2.84), log turnover (+1.00)
high leverage · cheap (high B/P) · high cash yield — 208 names; mostly Industrials; drivers: leverage debt to mcap (+3.01), val bp z (+0.70), val cfop z (+0.40)
expensive (low E/P) · high leverage · low margin — 204 names; mostly Real Estate; drivers: val ep z (−3.89), leverage debt to mcap (+2.57), profit margin (−1.54)
high volatility (3m) · rising (3m) · strong 1y momentum — 196 names; mostly Industrials; drivers: vol of vol 63 (+5.00), rv 63 (+4.98), ret 126d (+4.84)
high downside vol · high volatility (3m) · high volatility (1y) — 167 names; mostly Consumer Cyclical; drivers: vol of vol 63 (+4.94), downside vol 63 (+4.63), rv 63 (+4.48)
strong 1y momentum · wide 1y upside range · favorable 1y edge — 161 names; mostly Industrials; drivers: ret 252d (+2.91), pa mfe 52w (+2.85), ret 126d (+2.41)
rising (3m) · high volatility (3m) · strong 1y momentum — 66 names; mostly Consumer Cyclical; drivers: vol of vol 63 (+4.90), ret 20d (+4.90), ret 63d (+4.89)

Machine-readable data (free, read-only JSON)

The full map, per-ticker cluster assignments with confidence and anomaly scores, and PCA structure are published as open JSON for automated and AI-analyst consumption:

/api/ml/latest?market=hk — clusters, names, sector mix, representative tickers, top features
/api/ml/tickers?market=hk — per-ticker cluster, confidence, anomaly score, day-over-day migration, tenure
/api/ml/pca?market=hk — PCA components, explained variance, feature loadings

Descriptive market-structure research only. Unsupervised clustering finds structure, not direction; a tight cluster or an anomaly is a starting point for research, never a trade signal.