qbuntuqbuntu

宮越ホールディングス (6620.T) quant model fit

6620.T·不動産·jp-listed
computed asof 2026-10-01T02:33:09

Median hit rate and return when a model selected 宮越ホールディングス (6620.T). Model: ml-factor-lgbm-9f-jp-growth. Model–stock fit, not a buy signal.

Median LONG hit rate is 0.32 (n=12). below a coin flip.Median LONG return is -238.6 bps.

read this first — not a buy signal

Model predictability decomposition: measures how accurately the ml-factor-lgbm-9f-jp-growth model family predicts this stock's returns within the jp-listed universe. High hit_rate_long does NOT mean 'good stock to buy' -- it means 'the model historically predicted this stock well when selecting it LONG'. This is a property of model-stock fit, not intrinsic stock quality.

Decomposes ml-factor-lgbm-9f-jp-growth — strategy verdict: marginal. Standalone walk-forward run of the same LightGBM 9-factor family on the jp-listed universe (~4,100 tickers); the referenced record's verdict applies to the jp-growth universe only.

Accuracy (median across grid)

metricmedianp25 – p75n
Hit rate · LONG0.31990.2580 – 0.358012
Hit rate · SHORT0.40060.3748 – 0.440612
Avg return · LONG (bps)-238.6-273.0 – -137.712
Avg return · SHORT (bps)30.0-129.0 – 79.812
Selection rate · LONG0.00430.0034 – 0.005612
Selection rate · SHORT0.00340.0031 – 0.004212

Hit rate ≈ 0.5 means the model's picks on this stock were near coin-flip. Returns in bps per trade (1 bps = 0.01%). Metrics with n = 0 are omitted.

PnL contribution

long
-73.7
short
-0.1981
total
-73.9

Selected LONG 3335 / SHORT 2743 times out of 747,264 appearances (0.446% LONG).

sum of per-trade decimal returns across all grid combos and retrain periods (0.01 = one trade returning 1%)

/ai/stocks/6620.T.md/v1/stocks/6620.T
Research, not investment advice. Metrics describe model–stock fit, not the stock's merit.