gavindiaz/mql5
1
0
Fork 0
mirror of https://github.com/softwaredevelop/mql5.git synced 2026-11-02 11:16:15 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
main
mql5/Indicators/MyIndicators/ChaosWeightedRSI_Pro_Documentation.md
T
Toh4iem9 810d766191 docs: add document
2026-10-25 10:21:11 +01:00

17 KiB
Raw Permalink Blame History

Institutional Chaos-Weighted RSI Pro (Enterprise Ultra Edition v2.50)

1. Executive Summary & Microstructure Foundations

The Institutional Chaos-Weighted RSI Pro (ChaosWeightedRSI_Pro) is an institutional-grade adaptive momentum oscillator developed for MetaTrader 5 (Year 2026 Enterprise Edition). It resolves the classic limitations of J. Welles Wilder's Relative Strength Index (RSI, 1978) by dynamically modulating the internal smoothing factor using non-linear Fractal Dimension (Chaos) Metrics.

The Flaw of the Classical Static RSI

Wilder's traditional RSI employs a static smoothing constant:

\alpha_{\text{base}} = \frac{1}{\text{Length}}

In electronic liquidity pools, this static factor creates an unavoidable structural dilemma:

  • During low-volatility, chaotic consolidation: The static alpha overreacts to high-frequency micro-whipsaws, generating frequent false overbought/oversold extremes and premature counter-trend signals.
  • During explosive directional trends: The static alpha rapidly saturates into extreme territory (>70 or <30), misleading traders into exiting winning runners prematurely.

The Adaptive Chaos-Dampening Solution

ChaosWeightedRSI_Pro bridges fractal geometry with momentum:

  • It continuously measures the Fractal Dimension Index (FDI) of the price path via the central CFractalEngine.
  • When the market enters a chaotic, noisy regime (Chaos \to 1.0), the smoothing alpha decays toward zero (\alpha_{\text{adj}} \to 0.001). The oscillator dampens and freezes, refusing to generate false whipsaws.
  • When the market enters a clean, persistent trend (Chaos \to 0.0), the smoothing alpha expands back to its maximum velocity, tracking directional order flow with instantaneous responsiveness.

2. Mathematical & DSP Engine Specifications

The calculation engine CChaosWeightedRSI_Calculator operates on chronological arrays (ArraySetAsSeries(..., false)) and embeds CFractalEngine alongside CHeikinAshi_Calculator.


+-------------------------------------------------------------------------------+
|                       CHAOS-WEIGHTED RSI CALCULATION PIPELINE                 |
+-------------------------------------------------------------------------------+
  Price Stream: P_t (Standard or Heikin Ashi via ENUM_APPLIED_PRICE_HA_ALL)
         |
         v
  Central CFractalEngine:
    Compute Sevcik Path Length L over InpDisorderLength (default: 20)
    FDI_t = 1.0 + ln(L) / ln(2*(N-1))
         |
         v
  Normalized Chaos Ratio (Chaos_t in [0.0, 1.0]):
    Chaos_t = Clamp( (FDI_t - 1.0) / 0.5, 0.0, 1.0 )
         |
         v
  Modulated Smoothing Factor (alpha_adj):
    alpha_adj = Clamp( alpha_base* (1.0 - Chaos_t)^Sensitivity, 0.001, 1.0 )
         |
         v
  Pipelined Adaptive RMA Accumulators:
    AvgGain_t = AvgGain_{t-1} + alpha_adj *(Gain_t - AvgGain_{t-1})
    AvgLoss_t = AvgLoss_{t-1} + alpha_adj* (Loss_t - AvgLoss_{t-1})
         |
         v
  Chaos-Weighted RSI Output (WRSI_t in [0.0, 100.0]):
    RS = AvgGain_t / AvgLoss_t  -->  WRSI = 100.0 - (100.0 / (1.0 + RS))
         |
         v
  Unified Moving Average Engine (Optional Signal Line via MovingAverage_Engine.mqh)

A. The Chaos-Modulated Smoothing Equation

The base smoothing alpha is derived from the user-defined RSI length:

\alpha_{\text{base}} = \frac{1}{\text{InpRSILength}}

The non-linear modulation equation adjusts the responsiveness based on market entropy:

\alpha_{\text{adj}} = \max\Big( 0.001, \, \min\big( 1.0, \, \alpha_{\text{base}} \cdot (1.0 - Chaos_t)^{\text{InpSensitivity}} \big) \Big)
  • In a Clean Trend (Chaos = 0.0): (1.0 - 0.0)^{\text{Sens}} = 1.0 \implies \alpha_{\text{adj}} = \alpha_{\text{base}}. Full momentum tracking.
  • In Extreme Chaos (Chaos = 1.0): (1.0 - 1.0)^{\text{Sens}} = 0.0 \implies \alpha_{\text{adj}} = 0.001. The recursive filter locks in place, acting as an automatic whipsaw brake.

B. Gain and Loss Formulation

\text{Change}_t = P_t - P_{t-1} \text{Gain}_t = \max(\text{Change}_t, \, 0.0), \quad \text{Loss}_t = \max(-\text{Change}_t, \, 0.0) \text{AvgGain}_t = \text{AvgGain}_{t-1} + \alpha_{\text{adj}} \cdot (\text{Gain}_t - \text{AvgGain}_{t-1}) \text{AvgLoss}_t = \text{AvgLoss}_{t-1} + \alpha_{\text{adj}} \cdot (\text{Loss}_t - \text{AvgLoss}_{t-1}) \text{WRSI}_t = \begin{cases} 100.0 & \text{if } \text{AvgLoss}_t \le 1.0 \times 10^{-12} \\ 0.0 & \text{if } \text{AvgGain}_t \le 1.0 \times 10^{-12} \\ 100.0 - \left( \frac{100.0}{1.0 + \frac{\text{AvgGain}_t}{\text{AvgLoss}_t}} \right) & \text{otherwise} \end{cases}

3. Centralized CFractalEngine Integration

In accordance with Level 5 Enterprise Architecture conventions (Single Responsibility Principle & DRY), CChaosWeightedRSI_Calculator does not duplicate fractal dimension mathematics. Instead, it embeds <MyIncludes\Fractal_Engine.mqh>:

  • The fractal engine precalculates coordinate invariants (\Delta x)^2 and denominator logs in Init().
  • On each bar, m_fractal_engine.CalculateChaosPoint() is evaluated in O(\text{DisorderLength}) time.
  • Pipelined reciprocal multiplication (inv\_range = 1.0 / \text{Range}) eliminates runtime divisions.

4. Non-Repainting Divergence Engine & Subwindow Trendlines

The indicator includes a strictly non-repainting Subwindow Divergence Engine that detects structural discrepancies between price extremes and the Chaos-Weighted RSI:


Regular Bullish Divergence (Subwindow Trendline: clrMediumSeaGreen, Dotted)
  Price Action:   Lower Low (P_2 < P_1)
  Chaos RSI:      Higher Low (WRSI_2 > WRSI_1)
  Interpretation: Selling momentum is exhausted despite lower price. Imminent reversal up.

Regular Bearish Divergence (Subwindow Trendline: clrCrimson, Dotted)
  Price Action:   Higher High (P_2 > P_1)
  Chaos RSI:      Lower High (WRSI_2 < WRSI_1)
  Interpretation: Buying momentum is exhausted despite higher price. Imminent reversal down.

The p + d Non-Repainting Protocol

  • A pivot at bar p requires d confirmation bars to the left and d bars to the right (InpDivStrength).
  • Therefore, a pivot is permanently confirmed and locked at bar p + d.
  • Once confirmed, a trendline is drawn connecting the confirmed pivot points in the subwindow (OBJ_TREND, Style: STYLE_DOT).
  • Zero Repaint Guarantee: The trendline coordinates are anchored to historical timestamps. They will never move or disappear on future ticks.

5. High-Performance 2026 MTF Fast-Path & Multi-Instance Architecture

A. True MTF Divergences (Eliminating the Stepped-Curve Paradox)

When viewing higher timeframe CWRSI on an execution chart (e.g., viewing H1 CWRSI on an M15 chart):

  • Conventional indicators run divergence checks on the stepped lower timeframe curve, resulting in false triggers on flat steps.
  • ChaosWeightedRSI_Pro evaluates divergences strictly in the clean, unstepped HTF domain (h_time[], h_high[], h_low[], h_res_wrsi[]).
  • Confirmed HTF divergence trendlines are anchored directly to the higher timeframe candle open timestamps (h_time), projecting clean, mathematically pure macro divergence lines onto the lower timeframe chart.

B. Multi-Instance Namespace Isolation (InpInstanceID)

To enable multiple indicator instances on a single chart (e.g., H1 CWRSI in Subwindow 1 and H4 CWRSI in Subwindow 2):

  • Every graphical object is prefixed with a deterministic instance-isolated token:

    g_prefix = StringFormat("CWRSI_%I64d_%s_%d_%d_%d_",
                        ChartID(),
                        EnumToString(g_calc_timeframe),
                        InpRSILength,
                        InpDisorderLength,
                        InpInstanceID);

  • Different instances will never delete, overwrite, or collide with each other's graphical objects.

  • GetIndicatorSubwindow() automatically binds each instance's trendlines to its own designated subwindow.

6. Comprehensive Parameter Reference & Calibration Matrix

Input Settings

Parameter Type Default Description Operational Profile
InpTimeframe ENUM_TIMEFRAMES PERIOD_CURRENT Calculation timeframe (Current or HTF). Native O(1) or Two-Pointer MTF.
InpRSILength int 14 Base RSI analysis window length. Governs momentum lookback.
InpDisorderLength int 20 Fractal Dimension (Chaos) lookback window. Governs noise filter sensitivity.
InpSensitivity double 1.0 Chaos dampening exponent (0.1–5.0). Higher = more aggressive freezing in chop.
InpPriceSource ENUM_APPLIED_PRICE_HA_ALL PRICE_CLOSE_STD Applied price source (Standard / Heikin Ashi). Full Heikin Ashi support.
InpHistoryBars int 1000 Historical lookback limit (0 = Full Chart). Eliminates cold-start freezes.
InpColorRSI color clrDodgerBlue Monochrome CWRSI line color. Primary indicator plot.
InpWidthRSI int 2 CWRSI line thickness. Visual width.
InpShowSignal bool false Display Moving Average Signal Line. Optional momentum overlay.
InpSignalPeriod int 9 Signal Line smoothing period. Governs signal responsiveness.
InpSignalType ENUM_MA_TYPE EMA Signal Line MA Type (SMA, EMA, SMMA, etc.). Unified Moving Average Engine.
InpColorSignal color clrMagenta High-contrast Signal Line color. Universal contrast on light/dark themes.
InpWidthSignal int 2 Signal Line thickness. Visual width.
InpShowDivLines bool true Display subwindow divergence trendlines. Toggles graphical lines.
InpDivStrength int 15 Divergence pivot confirmation strength (d). Bars required left and right.
InpBullDivColor color clrMediumSeaGreen Bullish divergence trendline color. Visual trendline styling.
InpBearDivColor color clrCrimson Bearish divergence trendline color. Visual trendline styling.
InpInstanceID int 1 Unique instance identifier for multi-setups. Prevents multi-indicator collisions.

Quantitative Calibration Matrix

Trading Style Asset Class Base Chart Target TF RSI / FDI Length Sensitivity Price Source Strategy Focus
Micro-Scalping Crypto / Indices M1 / M3 PERIOD_CURRENT 9 / 14 1.2 PRICE_MEDIAN_STD Ultra-fast momentum with tight noise freezing during order book sweeps.
Day Trading Majors FX / Metals M5 / M15 M15 / H1 14 / 20 1.0 PRICE_CLOSE_STD Standard institutional calibration for session momentum and divergence fades.
Swing Structure Equities / Crude Oil H1 / H4 H4 / D1 14 / 24 1.5 PRICE_HA_CLOSE Macro trend surfing filtered by Heikin Ashi smoothing to eliminate daily wicks.

7. Quantitative Synergy Playbooks

Playbook A: Chaos-Filtered Trend Momentum (RSI Cross 50)

Price Action
  |                         [ACCELERATING TREND BREAKOUT] /=====> STRONG BUY ENTRY
  |                                                      /
  |-----------------------------------------------------/------------------------------ Resistance
CWRSI Subwindow
  | ----------------------------------------------------------------------------------- Level 70.0 (Overbought)
  |                              /======================> WRSI SURGES ABOVE 50.0
  | ============================/====================================================== Level 50.0 (Centerline)
  |                            ^
  |                            |-- BULLISH MOMENTUM CROSS (Market Order Confirmed)
  | ----------------------------------------------------------------------------------- Level 30.0 (Oversold)
  1. Setup:
    • Market breaks out of a consolidation pattern.
    • CWRSI breaks cleanly above the 50.0 median centerline.
    • Diagnosis: Because the indicator dynamically dampens in noise, a cross above 50 occurs only when the trend has sufficient directional efficiency.
  2. Execution Trigger:
    • Enter Long on the bar close confirming BufferWRSI[1] > 50.0.
    • (Optional: If InpShowSignal = true, verify BufferWRSI[1] > BufferSignal[1]).
  3. Exit: Hold trade while BufferWRSI remains above 50.0. Exit on a close back below 50.0 or upon divergence formation.

Playbook B: Non-Repainting Macro Divergence Reversal

  1. Setup:
    • Load ChaosWeightedRSI_Pro on an M15 chart with InpTimeframe = PERIOD_H1 and InpShowDivLines = true.
    • Price prints a higher high on the chart, but a dashed red trendline (clrCrimson) appears in the subwindow connecting a lower high on the H1 CWRSI.
  2. Execution Trigger:
    • Wait for an M15 reversal candle (e.g., Bearish Engulfing or rejection wick) confirming the H1 macro divergence.
    • SELL: Enter Short targeting the opposing H1 pivot or the 30.0 oversold level.
  3. Risk Management:
    • Stop Loss: 5 ticks above the H1 swing high.

Playbook C: The Quad-Suite Gated Alignment

Combine all four core suite engines for institutional execution:

  1. Regime State (Entropy_Pro): SampEn < 0.80 (Structural order confirmed).
  2. Volatility Trigger (VolatilityRegime_Pro): Ratio surges from Blue (< 0.75) into Lime (> 1.00).
  3. Adaptive Momentum (ChaosWeightedRSI_Pro): CWRSI crosses above 50.0 with no opposing bearish divergence.
  4. Trend Geometry (KAMA_Bands_Pro): Price band-walks along +1.0\sigma.
  • Trade Quality: Institutional Grade A+ Setup.

8. Developer Integration & Expert Advisor Guide

The indicator exposes its visual buffers directly to MetaTrader 5 Expert Advisors (EAs) via the native iCustom() API:

Buffer Index Map

Buffer Index Identifier Description Range / Output
0 BufferWRSI Chaos-Weighted RSI value. Continuous double (0.0 \dots 100.0).
1 BufferSignal Signal Line (Unified Moving Average). Continuous double (0.0 \dots 100.0) or EMPTY_VALUE.

Example MQL5 EA Integration Snippet

//+------------------------------------------------------------------+
//| Using ChaosWeightedRSI_Pro in an Automated Expert Advisor (EA)   |
//+------------------------------------------------------------------+
int h_cwrsi = INVALID_HANDLE;

int OnInit()
  {
   // Bind to ChaosWeightedRSI_Pro via iCustom
   h_cwrsi = iCustom(_Symbol, PERIOD_CURRENT, "ChaosWeightedRSI_Pro",
                     PERIOD_CURRENT,         // Timeframe
                     14,                     // RSI Length
                     20,                     // Disorder Length (FDI)
                     1.0,                    // Sensitivity
                     PRICE_CLOSE_STD,        // Price Source
                     1000,                   // History Bars Limit
                     clrDodgerBlue,          // RSI Line Color
                     2,                      // RSI Line Width
                     true,                   // Enable Signal Line
                     9,                      // Signal Period
                     EMA,                    // Signal MA Type (ENUM_MA_TYPE)
                     clrMagenta,             // Signal Color
                     STYLE_SOLID,            // Signal Style
                     2,                      // Signal Width
                     false,                  // Disable GUI Trendlines in EA for speed!
                     15,                     // Divergence Strength
                     clrMediumSeaGreen,      // Bullish Div Color
                     clrCrimson,             // Bearish Div Color
                     1);                     // Instance ID

   if(h_cwrsi == INVALID_HANDLE)
     {
      Print("Critical Error: Failed to create handle for ChaosWeightedRSI_Pro!");
      return INIT_FAILED;
     }
   return INIT_SUCCEEDED;
  }

void OnTick()
  {
   double wrsi[2], signal_line[2];
   // Read bar 1 and bar 2 (completed closed bars) for non-repainting execution
   if(CopyBuffer(h_cwrsi, 0, 1, 2, wrsi) <= 0 ||
      CopyBuffer(h_cwrsi, 1, 1, 2, signal_line) <= 0)
      return;

   double wrsi_curr   = wrsi[1];
   double wrsi_prev   = wrsi[0];
   double signal_curr = signal_line[1];
   double signal_prev = signal_line[0];

   // RULE 1: MOMENTUM ACCELERATION CROSSOVER
   bool bull_cross = (wrsi_prev <= signal_prev && wrsi_curr > signal_curr && wrsi_curr > 50.0);
   bool bear_cross = (wrsi_prev >= signal_prev && wrsi_curr < signal_curr && wrsi_curr < 50.0);

   if(bull_cross)
     {
      // Chaos-filtered bullish momentum confirmed! Execute Long.
      ExecuteOrder(ORDER_TYPE_BUY);
     }
   else if(bear_cross)
     {
      // Chaos-filtered bearish momentum confirmed! Execute Short.
      ExecuteOrder(ORDER_TYPE_SELL);
     }
  }

*Enterprise Quantitative Software Suite — 2026 Ultra Edition

Reference in New Issue View Git Blame Copy Permalink