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
684fea7878d0e8466a2df4e5e879fdaf45c869df
mql5/Indicators/MyIndicators/DMIStochastic_Pro_Documentation.md
T
Toh4iem9 1256b56316 docs: refactor
2026-09-28 17:36:07 +02:00

21 KiB
Raw Blame History

DMI Stochastic Suite Pro (v3.10)

Barbara Star's Hybrid Directional Movement & Stochastic Momentum Oscillator

1. Summary (Introduction)

DMI Stochastic Pro (v3.10) is an institutional-grade hybrid momentum oscillator developed by Barbara Star, Ph.D. It uniquely synthesizes J. Welles Wilder's Directional Movement System (DMI) with George Lane's Stochastic Oscillator.

Standard Stochastic oscillators evaluate price directly relative to the high-low price range, which frequently generates severe whipsaws and premature overbought/oversold signals during sustained directional trends. DMI Stochastic resolves this by normalizing the Directional Movement Index (+DI minus -DI) through the Stochastic algorithm.

The result is a bounded (0 to 100) trend-momentum oscillator that filters out counter-trend noise, identifies high-conviction trend pullbacks, and pinpoints statistical exhaustion extremes with remarkable fidelity.

In high-density multi-chart workspaces—such as setups operating 14 active chart windows with MTF DMI Stochastic running in Subwindow 2—legacy implementations cause severe UI freezing due to repetitive iBarShift queries and tick-by-tick heap memory churn. Version 3.10 Enterprise Edition eliminates this latency via a Persistent Memory Architecture and Zero-Lag MTF Fast-Path.


┌────────────────────────────────────────────────────────────────────────┐
│               DMI STOCHASTIC ARCHITECTURAL EVOLUTION                   │
├────────────────────────────────────────────────────────────────────────┤
│                                                                        │
│   LEGACY ENGINE (v3.00):                                               │
│   ┌──────────────────────────────────────────────────────────────┐     │
│   │ • 5 separate Copy API calls executed on every MTF tick       │     │
│   │ • Up to 500 iBarShift API calls per tick in forming blocks   │     │
│   │ • Dynamic heap reallocations (vol_double[]) on every tick    │     │
│   │ • 14 Charts × Subwindow 2 = Massive UI Thread Freezing       │     │
│   └──────────────────────────────────────────────────────────────┘     │
│                                  │                                     │
│                                  ▼ OPTIMIZED                           │
│   ENTERPRISE ENGINE (v3.10):                                           │
│   ┌──────────────────────────────────────────────────────────────┐     │
│   │ • 1 Atomic CopyRates call per live MTF tick (-80% API calls) │     │
│   │ • 0 iBarShift calls on live ticks (ArrayBsearch Fast-Path)   │     │
│   │ • Persistent Member Buffers (Zero heap allocations on ticks) │     │
│   │ • Direct Fused DMI Integration (Precomputed RMA math)        │     │
│   └──────────────────────────────────────────────────────────────┘     │
│                                                                        │
└────────────────────────────────────────────────────────────────────────┘

Key Capabilities

  • Hybrid Trend-Momentum Synthesis: Normalizes directional force (+DI vs. -DI) into a standardized 0–100 bounded scale.
  • Persistent Memory Pipeline: Eliminates dynamic array allocations on live ticks, ensuring zero garbage-collection pauses.
  • Fused DMI Engine Composition: Inherits the v1.30 DMI_Engine optimizations, utilizing precomputed Wilder multipliers and fused single-pass smoothing.
  • Zero-Lag MTF Fast-Path: Higher-timeframe curves (e.g., M5 DMI Stoch on M1 charts) project as non-warping steps via DataSync_Tools.mqh with zero iBarShift overhead on live ticks.
  • Selectable Smoothing Algorithms: Supports 8 moving average algorithms for %K slowing and %D signal lines (SMA, EMA, SMMA, LWMA, TMA, DEMA, TEMA, VWMA).

2. Mathematical Foundations & DMI-Stochastic Hybridization


     Raw Price (OHLC) ──▶ [ Fused DMI Engine ] ──▶ +DI(t) and -DI(t)
                                                    │
                                                    ▼
                       DMI Oscillator = +DI(t) - -DI(t)
                                                    │
                                                    ▼
                       Stochastic Normalization (0 - 100)
                                                    │
                                 ┌──────────────────┴──────────────────┐
                                 ▼                                     ▼
                           Fast %K (Raw)                      Slow %K & Signal %D

2.1. DMI Differential Oscillator (\text{DmiOsc}_t)

First, the indicator computes the Directional Indicators (+\text{DI}_t and -\text{DI}_t) using Welles Wilder's True Range and Directional Movement formulas across period P_{\text{DMI}} = \text{InpDMIPeriod}:

\text{DmiOsc}_t = \begin{cases} (+\text{DI}_t) - (-\text{DI}_t), & \text{if } \text{InpOscType} = \text{OSC\_PDI\_MINUS\_NDI} \\ (-\text{DI}_t) - (+\text{DI}_t), & \text{if } \text{InpOscType} = \text{OSC\_NDI\_MINUS\_PDI} \end{cases}

The resulting oscillator oscillates around an equilibrium zero-line within theoretical bounds of [-100, +100].

2.2. Stochastic Normalization Kernel (Fast %K)

The Stochastic algorithm is then applied directly to the \text{DmiOsc} series across a lookback window K_{\text{fast}} = \text{InpFastKPeriod}:

\text{HighestOsc}_t = \max_{j=0 \dots K_{\text{fast}}-1} (\text{DmiOsc}_{t-j}) \text{LowestOsc}_t = \min_{j=0 \dots K_{\text{fast}}-1} (\text{DmiOsc}_{t-j}) \text{Range}_t = \text{HighestOsc}_t - \text{LowestOsc}_t \text{FastK}_t = \begin{cases} \frac{\text{DmiOsc}_t - \text{LowestOsc}_t}{\text{Range}_t} \cdot 100, & \text{if } \text{Range}_t > 10^{-9} \\ 50.0, & \text{otherwise} \end{cases}

2.3. Slow %K and %D Signal Smoothing Equations

To eliminate high-frequency whipsaws, \text{FastK} is smoothed using the selected InpStochMethod across K_{\text{slow}} = \text{InpSlowKPeriod}:

\%K_t = \text{MovingAverage}(\text{FastK}, K_{\text{slow}}, \text{InpStochMethod})

The %D signal line is generated by smoothing \%K across period D_{\text{smooth}} = \text{InpSmoothPeriod} using InpSignalMethod:

\%D_t = \text{MovingAverage}(\%K, D_{\text{smooth}}, \text{InpSignalMethod})

2.4. Symmetrical 5-Zone Indicator Level Matrix

Level Threshold Level Name Market State Institutional Action
90.0 Extreme Overbought Parabolic buying climax; exhaustion imminent. Tighten trailing stops; avoid fresh breakout longs.
80.0 Overbought Warning Strong bullish trend momentum active. Bullish continuation zone; trail stops behind %K.
50.0 Equilibrium Center Zero-bias inflection line (+DI \approx -DI). Directional pivot: >50 Bullish bias, <50 Bearish bias.
20.0 Oversold Warning Strong bearish trend momentum active. Bearish continuation zone; trail stops above %K.
10.0 Extreme Oversold Capitulation selling floor; short covering risk. Prepare for mean-reversion bounce; cover short positions.

3. MQL5 Architecture & Computational Benchmarking


┌────────────────────────────────────────────────────────┐
│                     DMI_Engine.mqh                     │
│    (Fused TR/DM Smoothing + Normalized DI in 1 Pass)   │
└──────────────────────────┬─────────────────────────────┘
                           │ Delivers +DI and -DI in O(1)
                           ▼
┌────────────────────────────────────────────────────────┐
│              DMIStochastic_Calculator.mqh              │
│  (Persistent Memory Pipeline: Zero Dynamic Allocation) │
└──────────────────────────┬─────────────────────────────┘
                           │ Delivers %K and %D in O(1)
                           ▼
┌────────────────────────────────────────────────────────┐
│                 DMIStochastic_Pro.mq5                  │
│       (Unified Native & Zero-Lag MTF Fast-Path)        │
├──────────────────────────┬─────────────────────────────┤
│   Buffer Layer (2)       │   Centralized Framework     │
│   • BufferK (Plot 1: %K) │   • DataSync_Tools.mqh      │
│   • BufferD (Plot 2: %D) │   • Atomic CopyRates MTF    │
│                          │   • Binary Search Snapping  │
└──────────────────────────┴─────────────────────────────┘

Engineering Benchmark: Legacy (v3.00) vs. Enterprise (v3.10)

Metric Legacy Implementation (v3.00) Enterprise Refactor (v3.10) Net Optimization
MTF Live-Tick iBarShift Up to 500 calls per tick 0 calls on live ticks (ArrayBsearch) Complete Zero-Lag
MTF Data Copy Calls 5 separate calls per tick 1 atomic CopyRates query -80.0% API Overhead
Heap Memory Churn Dynamic vol_double[] every tick Persistent Member Cache Zero Allocation Pauses
Pointer Safety Slow CheckPointer() on every tick Fast if(!g_calculator) Guard Optimized Branching
Multi-Window Scalability Severe UI lag on 14 charts Silky-smooth execution on >14 charts Enterprise Certified

4. Parameters Reference

Timeframe Settings

  • InpTimeframe (default: PERIOD_CURRENT): Calculation timeframe. Set to PERIOD_CURRENT for native zero-lag execution, or choose a higher timeframe (e.g., PERIOD_M5, PERIOD_H1) to activate the synchronized MTF engine.

DMI & Stochastic Settings

  • InpCandleSource (default: CANDLE_STANDARD): Input price series source (CANDLE_STANDARD or CANDLE_HEIKIN_ASHI).
  • InpOscType (default: OSC_PDI_MINUS_NDI): Differential formula (OSC_PDI_MINUS_NDI for standard +DI - -DI orientation; OSC_NDI_MINUS_PDI for inverted view).
  • InpDMIPeriod (default: 10): Lookback period for underlying DMI +DI and -DI calculation.
  • InpFastKPeriod (default: 10): Stochastic lookback window for finding Highest/Lowest of the DMI Oscillator.
  • InpSlowKPeriod (default: 3): Smoothing period to transform Fast %K into Slow %K.
  • InpStochMethod (default: SMA): Moving average algorithm applied to %K (Supports SMA, EMA, VWMA, etc.).
  • InpSmoothPeriod (default: 3): Smoothing period to generate the %D signal line from %K.
  • InpSignalMethod (default: SMA): Moving average algorithm applied to %D.

Indicator Levels (0-100 Range)

  • InpLevelExtrHigh (default: 90.0): Extreme Overbought threshold.
  • InpLevelHigh (default: 80.0): Standard Overbought Warning threshold.
  • InpLevelMid (default: 50.0): Equilibrium Baseline.
  • InpLevelLow (default: 20.0): Standard Oversold Warning threshold.
  • InpLevelExtrLow (default: 10.0): Extreme Oversold threshold.
  • InpLevelColor (default: clrSilver): Color of horizontal level lines.
  • InpLevelStyle (default: STYLE_DOT): Line style of horizontal level lines.

Visual Settings

  • InpColorK / InpStyleK / InpWidthK: Visual styling for the %K main line (Default: clrDodgerBlue, STYLE_SOLID, 2).
  • InpColorD / InpStyleD / InpWidthD: Visual styling for the %D signal line (Default: clrCoral, STYLE_SOLID, 1).

5. Quantitative Trading Playbooks


┌────────────────────────────────────────────────────────────────────────┐
│                   DMI STOCHASTIC QUANTITATIVE PLAYBOOKS                │
├────────────────────────────────────────────────────────────────────────┤
│ 1. Trend Continuation Pullback: Buy pullbacks when %K bounces off the  │
│                                 50 or 20 level in an established trend.│
│ 2. Parabolic Exhaustion Fade:   Fade extreme climaxes when %K > 90     │
│                                 and crosses below %D.                  │
│ 3. Regime Inflection Cross:     Enter structural trend shifts when %K  │
│                                 cleanly crosses the 50.0 center-line.  │
└────────────────────────────────────────────────────────────────────────┘

5.1. Trend Continuation Pullback (The Institutional Edge)

  • Bullish Trend Setup:
    1. Market is in an established uptrend (price trading above SuperSmoother / VWAP).
    2. DMI Stochastic pulls back downward, dipping into the 50.0 equilibrium level or momentarily testing the 20.0 oversold level.
    3. %K crosses back strictly above %D while holding above 20.0 \rightarrow Enter Long. This represents a high-probability continuation entry where directional selling has exhausted.
  • Bearish Trend Setup:
    1. Market is in a confirmed downtrend.
    2. %K rallies upward to test the 50.0 or 80.0 resistance levels.
    3. %K crosses strictly below %D \rightarrow Enter Short.

5.2. Parabolic Climax & Capitulation Reversal

  • Extreme Bullish Exhaustion Fade:
    • %K expands beyond 90.0 (InpLevelExtrHigh), indicating extreme directional overextension.
    • %K rolls over and crosses below %D while inside the >80 zone \rightarrow Take profit on longs; initiate mean-reversion counter-trend shorts targeting the 50.0 median.
  • Extreme Bearish Capitulation Bounce:
    • %K flushes below 10.0 (InpLevelExtrLow), signaling forced liquidation selling.
    • %K hooks upward and crosses above %D \rightarrow Cover short positions; enter long targeting the 50.0 equilibrium.

5.3. Multi-Timeframe Alignment (M5 MTF DMI Stoch on M1 Execution)

  • Load DMIStochastic_Pro with InpTimeframe = PERIOD_M5 onto an M1 execution chart.
  • The non-warping flat staircase steps represent the 5-minute directional momentum:
    • If M5 %K is holding above 50.0: Focus exclusively on Long execution on M1.
    • If M5 %K is holding below 50.0: Focus exclusively on Short execution on M1.

6. Indicator Buffer Map (For Developers & EA Integration)

Buffer Allocation

Buffer Index Name Type Visual Plot Description
0 BufferK INDICATOR_DATA Plot 1 (DRAW_LINE) Main DMI Stochastic %K line.
1 BufferD INDICATOR_DATA Plot 2 (DRAW_LINE) Smoothed Signal %D line.

All buffers strictly maintain non-series chronological order (ArraySetAsSeries = false), ensuring direct compatibility with automated Expert Advisors via iCustom().

MQL5 EA Integration Interface Template

//+------------------------------------------------------------------+
//|                               EA_DMI_Stochastic_Interface.mq5    |
//|                                          Copyright 2026, xxxxxxxx|
//+------------------------------------------------------------------+
#property copyright "Copyright 2026, xxxxxxxx"
#property version   "1.00"
#property strict

//--- Include Calculator Definitions
#include <MyIncludes\DMIStochastic_Calculator.mqh>

//--- EA Inputs
input group "=== DMI Stochastic Filter Parameters ==="
input ENUM_TIMEFRAMES InpDMIStochTF    = PERIOD_CURRENT;    // Timeframe
input int             InpDMIPeriod     = 10;                // DMI Period
input int             InpFastKPeriod   = 10;                // Fast %K Period
input int             InpSlowKPeriod   = 3;                 // Slow %K Period
input int             InpSmoothPeriod  = 3;                 // %D Period
input ENUM_MA_TYPE    InpMethod        = SMA;               // Smoothing Method

//--- Global Indicator Handle
int g_dmistoch_handle = INVALID_HANDLE;

//+------------------------------------------------------------------+
//| Expert initialization function                                   |
//+------------------------------------------------------------------+
int OnInit()
  {
   if(g_dmistoch_handle != INVALID_HANDLE)
      IndicatorRelease(g_dmistoch_handle);

   // Instantiate handle to DMIStochastic_Pro via iCustom
   g_dmistoch_handle = iCustom(_Symbol,
                               InpDMIStochTF,
                               "DMIStochastic_Pro",
                               InpDMIStochTF,
                               0,                   // Standard Candle Source
                               OSC_PDI_MINUS_NDI,   // Oscillator Formula
                               InpDMIPeriod,
                               InpFastKPeriod,
                               InpSlowKPeriod,
                               InpMethod,
                               InpSmoothPeriod,
                               InpMethod);

   if(g_dmistoch_handle == INVALID_HANDLE)
     {
      PrintFormat("EA Error: Failed to create handle for DMIStochastic_Pro. Error: %d", GetLastError());
      return INIT_FAILED;
     }

   Print("EA Success: DMIStochastic_Pro handle initialized successfully.");
   return INIT_SUCCEEDED;
  }

//+------------------------------------------------------------------+
//| Expert deinitialization function                                 |
//+------------------------------------------------------------------+
void OnDeinit(const int reason)
  {
   if(g_dmistoch_handle != INVALID_HANDLE)
     {
      IndicatorRelease(g_dmistoch_handle);
      g_dmistoch_handle = INVALID_HANDLE;
     }
  }

//+------------------------------------------------------------------+
//| Expert tick function                                             |
//+------------------------------------------------------------------+
void OnTick()
  {
   // Query completed closed candle (Shift = 1) and previous candle (Shift = 2) for %K and %D
   double k_vals[2], d_vals[2];
   ArraySetAsSeries(k_vals, true); // Index 0 = Shift 1, Index 1 = Shift 2
   ArraySetAsSeries(d_vals, true);

   if(CopyBuffer(g_dmistoch_handle, 0, 1, 2, k_vals) < 2 ||
      CopyBuffer(g_dmistoch_handle, 1, 1, 2, d_vals) < 2)
     {
      return; // Data synchronizing
     }

   double k_bar1 = k_vals[0];
   double d_bar1 = d_vals[0];

   // Quantitative Momentum Signals
   bool is_overbought = (k_bar1 >= 80.0);
   bool is_oversold   = (k_bar1 <= 20.0);
   bool is_bullish    = (k_bar1 > 50.0);
   bool is_bearish    = (k_bar1 < 50.0);

   // Momentum Crossover Signals
   bool k_crossed_up   = (k_vals[1] <= d_vals[1] && k_vals[0] > d_vals[0]);
   bool k_crossed_down = (k_vals[1] >= d_vals[1] && k_vals[0] < d_vals[0]);

   // Telemetry Output
   Comment(StringFormat("DMI Stochastic Telemetry [Bar 1]:\n"
                        "%%K: %.2f | %%D: %.2f\n"
                        "Bias: %s | State: %s\n"
                        "Signals -> Buy: %s | Sell: %s",
                        k_bar1, d_bar1,
                        is_bullish ? "BULLISH (> 50.0)" : (is_bearish ? "BEARISH (< 50.0)" : "EQUILIBRIUM"),
                        is_overbought ? "OVERBOUGHT (>= 80)" : (is_oversold ? "OVERSOLD (<= 20)" : "NORMAL RANGE"),
                        (k_crossed_up && !is_overbought) ? "TRIGGERED (Bullish Cross)" : "NO",
                        (k_crossed_down && !is_oversold) ? "TRIGGERED (Bearish Cross)" : "NO"));
  }
//+------------------------------------------------------------------+
Reference in New Issue View Git Blame Copy Permalink