mirror of
https://github.com/mihakralj/QuanTAlib.git
synced 2026-08-17 01:58:06 +00:00
updates from mac
This commit is contained in:
Miha Kralj
@@ -0,0 +1,187 @@
|
||||
[](https://sonarcloud.io/summary/overall?id=mihakralj_QuanTAlib)
|
||||
[](https://app.codacy.com/gh/mihakralj/QuanTAlib/dashboard)
|
||||
[](https://codecov.io/gh/mihakralj/QuanTAlib)
|
||||
[](https://sonarcloud.io/summary/new_code?id=mihakralj_QuanTAlib)
|
||||
[](https://www.codefactor.io/repository/github/mihakralj/quantalib/overview/main)
|
||||
|
||||
[](https://www.nuget.org/packages/QuanTAlib/)
|
||||
|
||||
[](https://www.nuget.org/packages/QuanTAlib/)
|
||||
[](https://github.com/mihakralj/QuanTAlib/watchers)
|
||||
[](https://dotnet.microsoft.com/en-us/download/dotnet)
|
||||
|
||||
# QuanTAlib - Quantitative Technical Analysis Library
|
||||
|
||||
**Quan**titative **TA** **lib**rary (QuanTAlib) is a high-performance C# library for quantitative technical analysis, designed for [Quantower](https://www.quantower.com/) and other C#-based trading platforms.
|
||||
|
||||
## Key Features
|
||||
|
||||
- **Real-time streaming** - Indicators calculate results from incoming data without re-processing history
|
||||
- **Update/correction support** - Last value can be recalculated multiple times before advancing to next bar
|
||||
- **Valid from first bar** - Mathematically correct results from the first value with `IsHot` warmup indicator
|
||||
- **SIMD-optimized** - Hardware-accelerated vector operations (AVX/SSE) for batch processing
|
||||
- **Zero-allocation hot paths** - Minimal GC pressure for high-frequency scenarios
|
||||
|
||||
## Architecture
|
||||
|
||||
QuanTAlib uses a **Structure of Arrays (SoA)** memory layout optimized for numerical computing:
|
||||
|
||||
```
|
||||
┌─────────────────────────────────────────────────────────────┐
|
||||
│ Core Data Types │
|
||||
├─────────────────────────────────────────────────────────────┤
|
||||
│ TValue (16 bytes) │ Time-value pair (long + double) │
|
||||
│ TBar (48 bytes) │ OHLCV bar (long + 5 doubles) │
|
||||
│ TSeries │ Time series with SoA layout │
|
||||
│ TBarSeries │ OHLCV series with SoA layout │
|
||||
└─────────────────────────────────────────────────────────────┘
|
||||
|
||||
┌─────────────────────────────────────────────────────────────┐
|
||||
│ Data Feeds │
|
||||
├─────────────────────────────────────────────────────────────┤
|
||||
│ IFeed │ Unified feed interface │
|
||||
│ GBM │ Geometric Brownian Motion sim │
|
||||
│ CsvFeed │ CSV file reader │
|
||||
└─────────────────────────────────────────────────────────────┘
|
||||
|
||||
```
|
||||
|
||||
### Performance Design
|
||||
|
||||
The SoA layout stores timestamps and values in separate contiguous arrays:
|
||||
|
||||
```csharp
|
||||
// TSeries internal structure
|
||||
protected readonly List<long> _t; // Timestamps (contiguous)
|
||||
protected readonly List<double> _v; // Values (contiguous)
|
||||
|
||||
// Direct SIMD access via Span<T>
|
||||
ReadOnlySpan<double> values = series.Values;
|
||||
double avg = values.AverageSIMD(); // Hardware-accelerated
|
||||
```
|
||||
|
||||
This enables:
|
||||
- **Cache locality** - Sequential memory access patterns
|
||||
- **SIMD vectorization** - Process 4-8 values per CPU instruction
|
||||
- **Zero-copy access** - `CollectionsMarshal.AsSpan()` exposes internal arrays
|
||||
|
||||
## Quick Start
|
||||
|
||||
### Installation
|
||||
|
||||
```bash
|
||||
dotnet add package QuanTAlib
|
||||
```
|
||||
|
||||
### Basic Usage
|
||||
|
||||
```csharp
|
||||
using QuanTAlib;
|
||||
|
||||
// Create EMA indicator
|
||||
var ema = new Ema(period: 10);
|
||||
|
||||
// Streaming mode - process one value at a time
|
||||
TValue result = ema.Update(new TValue(DateTime.Now, price), isNew: true);
|
||||
|
||||
// Update current bar (e.g., price tick within same minute)
|
||||
result = ema.Update(new TValue(DateTime.Now, newPrice), isNew: false);
|
||||
|
||||
// Batch mode - process entire series
|
||||
var series = new TSeries();
|
||||
series.Add(prices); // Add historical data
|
||||
TSeries emaResults = Ema.Calculate(series, period: 10);
|
||||
```
|
||||
|
||||
### Multi-Period Analysis with SIMD
|
||||
|
||||
```csharp
|
||||
// Calculate multiple EMAs in parallel using SIMD
|
||||
int[] periods = { 9, 12, 26 };
|
||||
var emaVector = new EmaVector(periods);
|
||||
|
||||
// Single update calculates all periods
|
||||
TValue[] results = emaVector.Update(new TValue(time, price));
|
||||
Console.WriteLine($"EMA(9)={results[0]}, EMA(12)={results[1]}, EMA(26)={results[2]}");
|
||||
```
|
||||
|
||||
### Using Data Feeds
|
||||
|
||||
```csharp
|
||||
// Geometric Brownian Motion simulator
|
||||
var gbm = new GBM(startPrice: 100, mu: 0.05, sigma: 0.2);
|
||||
TBarSeries bars = gbm.Fetch(count: 1000, DateTime.UtcNow.Ticks, TimeSpan.FromMinutes(1));
|
||||
|
||||
// CSV file reader
|
||||
var csv = new CsvFeed("data/daily_IBM.csv");
|
||||
TBar bar = csv.Next(isNew: true);
|
||||
```
|
||||
|
||||
## Installation to Quantower
|
||||
|
||||
Copy DLL files to Quantower installation:
|
||||
|
||||
```
|
||||
<Quantower_root>\Settings\Scripts\Indicators\QuanTAlib\Averages\Averages.dll
|
||||
```
|
||||
|
||||
Where `<Quantower_root>` is the directory containing `Start.lnk`.
|
||||
|
||||
## Project Structure
|
||||
|
||||
```
|
||||
QuanTAlib/
|
||||
├── lib/
|
||||
│ ├── core/
|
||||
│ │ ├── tvalue/ # TValue struct
|
||||
│ │ ├── tseries/ # TSeries class
|
||||
│ │ ├── tbar/ # TBar struct
|
||||
│ │ ├── tbarseries/ # TBarSeries class
|
||||
│ │ └── simd/ # SIMD extensions
|
||||
│ ├── averages/
|
||||
│ │ └── ema/ # EMA indicator + tests + docs
|
||||
│ └── feeds/
|
||||
│ ├── csv/ # CSV file feed
|
||||
│ └── gbm/ # GBM simulator
|
||||
└── quantower/ # Quantower integration
|
||||
```
|
||||
|
||||
Each indicator follows a consistent file pattern:
|
||||
- `Indicator.cs` - Core implementation
|
||||
- `Indicator.Tests.cs` - Unit tests
|
||||
- `Indicator.Validation.Tests.cs` - Cross-validation with other libraries
|
||||
- `Indicator.md` - Documentation
|
||||
- `Indicator.Notebook.dib` - Interactive notebook
|
||||
- `Indicator.Quantower.cs` - Quantower wrapper
|
||||
|
||||
## Validation
|
||||
|
||||
QuanTAlib validates results against established TA libraries:
|
||||
|
||||
- [TA-LIB](https://www.ta-lib.org/function.html) - Industry standard C library
|
||||
- [Skender Stock Indicators](https://dotnet.stockindicators.dev/) - Popular .NET library
|
||||
- [Tulip Indicators](https://tulipindicators.org/) - High-performance C library
|
||||
|
||||
## Requirements
|
||||
|
||||
- .NET 8.0, 9.0, or 10.0
|
||||
- Hardware with AVX/SSE support recommended for optimal SIMD performance
|
||||
|
||||
## License
|
||||
|
||||
Apache License 2.0 - See [LICENSE](LICENSE) for details.
|
||||
|
||||
## Contributing
|
||||
|
||||
Contributions welcome! Each indicator should include:
|
||||
1. Core implementation with streaming support
|
||||
2. Unit tests covering edge cases
|
||||
3. Validation tests against reference libraries
|
||||
4. Documentation with mathematical formulas
|
||||
5. Quantower wrapper (optional)
|
||||
|
||||
## Links
|
||||
|
||||
- [GitHub Repository](https://github.com/mihakralj/QuanTAlib)
|
||||
- [NuGet Package](https://www.nuget.org/packages/QuanTAlib/)
|
||||
- [Quantower Platform](https://www.quantower.com/)
|
||||
Reference in New Issue
Block a user