Screens a 10-stock NSE basket (banks, IT, metals, higher-beta names) across roughly 15 technical indicators and scores current momentum on an additive 0–90 scale.
Computes each stock's historical forward 1/3/5-day return and win rate over a trailing 90-day window, counts bullish/bearish/neutral signals across seven independent checks, and combines all three into a normalized 0–100 composite ranking with an illustrative entry/stop/target plan for the top three.
Fixed in review
- A Parabolic SAR column that was NaN on every row, from subtracting two complementary, mutually-exclusive helper series.
- A technical-indicator heatmap mixing unnormalized, incompatible scales — RSI's 0–100 range visually dominating a 0/1 flag.
- Composite score weights that were not actually applying, because the three components were never normalized to comparable scales before weighting.
- Deprecated df.last('90D').
- An "Expected 3-Day Return" label that presented a purely historical average as if it were a forecast.
- A hardcoded future end date, and a momentum score labeled "0-100" whose true maximum was 90.
MomentumScreeningComposite scoringNSE