ANRF++（RSI+MACD+OBV+VR+CCL 自适应融合）· 不重绘

ANRF++ is a five-factor adaptive oscillator that fuses RSI, MACD, OBV, VR (Volume Ratio), and CCL (CLV→ADL momentum) into one clean signal. It’s designed to cut noise on lower timeframes while remaining responsive in trends.

How it works (simple)
Each factor is normalized to [-1, 1] (MACD/OBV/CCL use z-scores; RSI uses a centered scale; VR is mapped with a ratio transform).

A smooth tanh compression reduces spikes.

Adaptive weights are applied: weight ∝ |score|^p, then re-normalized.
→ Strong, consistent signals get more influence; weak/noisy signals get down-weighted.

The fused line (ANRF, range −1..1) is plotted with:

Deadband (entry filter)

Overbought/Oversold zones (±extLevel)

Signals
Long: ANRF crosses up through +deadband

Short: ANRF crosses down through −deadband

Optional markers when entering Overbought/Oversold.

Built-in alerts for all of the above.

Key Inputs
Factor windows: RSI length, MACD (fast/slow/signal), OBV Δ window, VR window, CCL window.

Fusion controls: base weights per factor, p-exponent, smoothing.

Deadband (signals) and Overbought/Oversold threshold (zones).

Optional display of component scores for tuning.

Notes
Non-repainting (no lookahead/security repainting).

Works on any market/timeframe; for scalping, use a higher deadband and larger p; for strong trends, lower them slightly.

Best used as a direction/quality filter alongside your execution rules (risk management required).