mirror of
https://github.com/softwaredevelop/mql5.git
synced 2026-07-27 20:47:44 +00:00
docs(indicators): refactor
This commit is contained in:
Toh4iem9
@@ -2,71 +2,86 @@
|
||||
|
||||
## 1. Summary (Introduction)
|
||||
|
||||
The `Market_Scanner_Pro` is a high-performance quantitative analysis tool designed to bridge the gap between technical charting and AI-assisted trading. It performs a multi-timeframe, multi-indicator scan across a portfolio of assets and exports the "Market State" into a structured CSV format suitable for Large Language Models (LLMs) or statistical analysis tools (Python/Excel).
|
||||
The `Market_Scanner_Pro` is a high-performance quantitative analysis tool designed to bridge the gap between technical charting and AI-assisted trading. It is an "Institutional Market X-Ray" that performs a multi-timeframe, multi-indicator scan across a portfolio of assets and exports the market state into a structured CSV format.
|
||||
|
||||
Instead of relying on basic price data, this script generates **"QuantScan 2.0"** metrics: it converts raw indicator values into normalized scores (e.g., Z-Score, Efficiency Ratio), providing a deep insight into Trend Quality, Momentum, and Statistical Extremes.
|
||||
This dataset ("QuantScan 3.0") is optimized for Large Language Models (LLMs) or statistical analysis tools. Instead of raw price data, it provides normalized scores (Z-Score, Efficiency Ratio, Relative Strength), offering deep insights into Trend Quality, Institutional Footprints, and Statistical Reversion risks.
|
||||
|
||||
## 2. Methodology and Logic
|
||||
|
||||
The script employs a **Hybrid Analysis Model**, splitting metrics into two logical timeframes:
|
||||
The script employs a **Hybrid Analysis Model** with three core layers:
|
||||
|
||||
1. **Context Layer (H1):** Analyzes the "Big Picture". It determines the dominant trend direction, the structural quality of that trend, and key support/resistance zones.
|
||||
2. **Trigger Layer (M15):** Analyzes the "Execution Timing". It looks for momentum shifts, volume anomalies, and statistical reversion signals.
|
||||
1. **Context Layer (H1):** Determines the "Big Picture". It identifies the dominant trend direction, the structural quality of that trend, and correlation with the broader market (Relative Strength).
|
||||
2. **Trigger Layer (M15):** Analyzes "Execution Timing". It monitors momentum shifts, volatility regimes, and statistical extremes.
|
||||
3. **Institutional Layer (New):** Detects hidden market mechanics, specifically "Absorption" (high volume vs. low range) and extreme probability of mean reversion.
|
||||
|
||||
### Key Metrics Calculated
|
||||
### Key Metrics Defined
|
||||
|
||||
* **Trend Score (Z-Score Proxy):** Measures how far the price is from the mean (DSMA) in units of volatility (ATR). A score of +2.0 means the price is 2 standard deviations above the trend.
|
||||
* **Trend Quality (Efficiency Ratio):** Differentiates between a smooth, tradeable trend (High ER) and a choppy, dangerous market (Low ER).
|
||||
* **Volatility Regime (Squeeze):** Identifies periods of low volatility (Bollinger Bands inside Keltner Channels) that often precede explosive moves.
|
||||
* **Volume Quality (RVOL):** Checks if the current move is supported by institutional volume (Relative Volume > 1.0).
|
||||
* **Trend Score (Z-Score & Deviation):** Measures how far the price is from the trend baseline in units of volatility (ATR).
|
||||
* **Relative Strength (RS):** Compares the asset's performance against a Benchmark (e.g., US500) over the last 24 hours. A positive RS indicates the asset is outperforming the market.
|
||||
* **Institutional Absorption:** A logical check based on Wyckoff principles. If Volume is extreme (RVOL > 2.0) but Price Movement is small, it indicates passive limit orders absorbing aggressive market orders—often a sign of a reversal.
|
||||
* **Reversion Probability:** A composite score (0-100%) that combines Z-Score extremes, Murrey Levels, and Momentum Exhaustion to predict a potential pullback.
|
||||
|
||||
## 3. MQL5 Implementation Details
|
||||
|
||||
The script is built upon the **"Professional Indicator Suite"** architecture, ensuring mathematical precision and performance.
|
||||
|
||||
* **Calculation Engines (`.mqh`):**
|
||||
Instead of using slow `iCustom` calls, the script directly instantiates the optimized Calculation Classes (e.g., `CDSMACalculator`, `CVWAPCalculator`) used by our indicators. This guarantees that the CSV data matches the chart visuals 100%.
|
||||
* **Defensive Programming:**
|
||||
The implementation includes rigorous "Safety Checks" (e.g., array bounds checking in ATR, data availability validation) to prevent runtime crashes, even when scanning hundreds of symbols.
|
||||
* **Smart Data Fetching:**
|
||||
It utilizes `FetchData` wrappers that efficiently retrieve OHLCV data and organize it into chronological arrays (`ArraySetAsSeries(false)`), optimized for our incremental calculation engines.
|
||||
* **Calculation Engines (`.mqh`):** It directly instantiates optimized Calculation Classes (e.g., `CDSMACalculator`, `CVWAPCalculator`) rather than using slow `iCustom` calls.
|
||||
* **Defensive Programming:** Includes rigorous safety checks (e.g., array bounds checking in ATR) to prevent runtime crashes during large-scale scanning.
|
||||
* **Smart Data Fetching:** Utilizes efficient `FetchData` wrappers to retrieve and sync OHLCV data for multiple timeframes instantaneously.
|
||||
|
||||
## 4. Parameters
|
||||
|
||||
* **Scanner Config:**
|
||||
* `InpUseMarketWatch`: If `true`, scans every active symbol in the Market Watch window.
|
||||
* `InpSymbolList`: A comma-separated list of symbols to scan if Market Watch is disabled (e.g., `EURUSD, BTCUSD, US500`).
|
||||
* `InpUseMarketWatch`: If `true`, scans all active symbols.
|
||||
* `InpSymbolList`: Custom symbol list (if using manual selection).
|
||||
* `InpBenchmark`: The symbol for Relative Strength comparison (Default: `US500`).
|
||||
* **`InpBrokerTimeZone`**: **NEW!** Your broker's timezone name (e.g. `EET`, `UTC+3`). This string is added to the CSV header so the AI knows the context of the timestamp (crucial for detecting Session Opens/Closes).
|
||||
* **`InpScanHistory`**: **NEW!** Number of bars to download for analysis (Default: `500`). Increase this if using slow moving averages (200 SMA).
|
||||
* **Timeframes:**
|
||||
* `InpTFFast`: The timeframe for Trigger metrics (Default: `M15`).
|
||||
* `InpTFSlow`: The timeframe for Context metrics (Default: `H1`).
|
||||
* **Metric Settings:**
|
||||
* Allows fine-tuning of indicators (e.g., `InpDSMAPeriod`, `InpLaguerreGamma`, `InpRVOLPeriod`).
|
||||
* `InpTFFast` (Trigger): Default `M15`.
|
||||
* `InpTFSlow` (Context): Default `H1`.
|
||||
* **Metric Settings:**
|
||||
* **`InpRSBars`**: **NEW!** Lookback period for Relative Strength calculation.
|
||||
* `24 (Default on H1)` = 24 Hours performance.
|
||||
* `120` = Weekly performance.
|
||||
* Indicators fine-tuning (DSMA, Gamma, etc).
|
||||
* **Squeeze Settings:**
|
||||
* Controls the sensitivity of the volatility squeeze detection (`BB Multiplier`, `KC Multiplier`).
|
||||
* Allows fine-tuning of the Volatility Squeeze sensitivity (`BB Multiplier`, `KC Multiplier`).
|
||||
* **TSI Settings:**
|
||||
* Customizable periods for the True Strength Index (Cycle).
|
||||
|
||||
## 5. Output Data Structure (CSV)
|
||||
## 5. Output Data Structure (CSV - QuantScan 3.0)
|
||||
|
||||
The script generates a file named `QuantScan_YYYY.MM.DD_HHMM.csv` in the `MQL5\Files` folder.
|
||||
|
||||
### Columns Explanation
|
||||
|
||||
| Header | Description | Interpretation |
|
||||
| Header | Description | Interpretation / ranges |
|
||||
| :--- | :--- | :--- |
|
||||
| **`TIME`** | Timestamp | `YYYY.MM.DD HH:MM` format. |
|
||||
| **`SYMBOL`** | Asset Name | e.g. `EURUSD`. |
|
||||
| **`SYMBOL`** | Asset Name | e.g., `EURUSD`. |
|
||||
| **`PRICE`** | Current Bid | The snapshot price at scan time. |
|
||||
| **`TREND_SCORE`** | **H1 Trend Strength** | Normalized deviation from trend. <br>• `> +1.0`: Strong Bull<br>• `< -1.0`: Strong Bear |
|
||||
| **`TREND_QUAL`** | **H1 Efficiency** | Quality of the trend (Kaufman ER). <br>• `> 0.6`: Clean Trend<br>• `< 0.3`: Noise/Chop |
|
||||
| **`ZONE`** | **H1 Structure** | Murrey Math Level. <br>• `Extreme`: Reversal likely.<br>• `Range`: Trading Zone. |
|
||||
| **`MOMENTUM`** | **M15 Laguerre** | Fast momentum (0.0 - 1.0). <br>• `> 0.8`: Bullish Pressure<br>• `< 0.2`: Bearish Pressure |
|
||||
| **`VOL_QUAL`** | **M15 RVOL** | Instant Institutional Interest. <br>• `> 1.5`: High Activity<br>• `< 0.8`: No interest |
|
||||
| **`TREND_SCORE`** | **H1 Trend Strength** | Normalized deviation. <br>• `> +1.0`: Strong Bull<br>• `< -1.0`: Strong Bear |
|
||||
| **`TREND_QUAL`** | **H1 Efficiency** | Trend noise filter (Kaufman ER). <br>• `> 0.6`: Clean Trend (Safe to trade) |
|
||||
| **`ZONE`** | **H1 Structure** | Murrey Math Level. <br>• `Extreme`: Reversal zone.<br>• `Range`: Trading zone. |
|
||||
| **`REL_STRENGTH`** | **Relative Perf.** | Performance vs Benchmark (24h). <br>• `> 0%`: Leader (Stronger than market)<br>• `< 0%`: Laggard (Weaker than market) |
|
||||
| **`MOMENTUM`** | **M15 Laguerre** | Fast momentum (0.0 - 1.0). <br>• `> 0.85`: Bullish Pressure (Gamma lag) |
|
||||
| **`VOL_QUAL`** | **M15 RVOL** | Relative Volume. <br>• `> 1.5`: High Activity<br>• `< 0.7`: Low Low Interest |
|
||||
| **`SQUEEZE`** | **M15 Vola State** | TTM Squeeze status. <br>• `ON`: Energy building (Prepare for breakout). |
|
||||
| **`TSI_DIR`** | **M15 Cycle** | True Strength Index direction (`BULL` / `BEAR`). |
|
||||
| **`Z_SCORE`** | **M15 Statistics** | Standard Deviations from mean. <br>• `> 2.5`: Statistically Extreme. |
|
||||
| **`VOL_REGIME`** | **M15 Vola Trend** | Ratio of Short/Long ATR. <br>• `> 1.0`: Volatility is expanding. |
|
||||
| **`TSI_DIR`** | **M15 Cycle** | Cycle direction (`BULL` / `BEAR`). |
|
||||
| **`REVERSION_PROB`** | **Reversion %** | Composite probability of a pullback. <br>• `> 80%`: High risk of reversal. |
|
||||
| **`ABSORPTION`** | **Inst. Volume** | Wyckoff Absorption signal. <br>• `YES`: High Vol + Small Body (Hidden activity). |
|
||||
|
||||
## 6. Usage Workflow
|
||||
|
||||
1. **Run the Script:** Drag `Market_Scanner_Pro` onto any chart.
|
||||
2. **Wait for Completion:** Check the "Experts" tab for progress. It usually takes a few seconds to scan 20-30 symbols.
|
||||
2. **Wait for Completion:** Check the "Experts" tab.
|
||||
3. **Locate File:** Open "File -> Open Data Folder -> MQL5 -> Files".
|
||||
4. **Process with AI:** Upload the `QuantScan_....csv` file to your LLM (GPT-4 / Claude 3) with a prompt like:
|
||||
> *"Analyze this market data. Identify high-quality trend setups where TREND_QUANT > 0.6 and SQUEEZE is ON. Also, warn me about mean reversion risks where Z_SCORE > 2.5."*
|
||||
4. **Process with AI:** Upload the `QuantScan_....csv` file to your LLM with a prompt like:
|
||||
|
||||
> *"Analyze this market data. Look for two specific setups:*
|
||||
>
|
||||
> 1. ***The Whale Utility:** Strong Trend (`TREND_SCORE > 0.5`) + Strong Relative Strength (`REL_STRENGTH > 0`) + Squeeze is `ON`.
|
||||
> 2. ***The Reversion Trap:** High Reversion Probability (`> 80%`) AND Absorption is `YES`.
|
||||
>
|
||||
> *List the top 3 candidates for each."*
|
||||
|
||||
Reference in New Issue
Block a user