DQuant Research is the open research hub of the DQuant ecosystem — a public space for papers, experiments, and practical notes on volatility forecasting and quantitative analysis.
Unlike the library documentation, Lab goes deeper: model comparisons, benchmarks on real market data, applied workflows, and methodology explained in plain language.
Live page: dquant.space/research
| Type | Description |
|---|---|
| Research | Formal studies with metrics, model comparisons, and statistical conclusions |
| Applied | Real-world use cases: workflows, tools, notebooks, and integrations |
| Methodology | Feature engineering, problem framing, common pitfalls, and best practices |
Materials are published in HTML (on the website) and Markdown (in this repository) when available, so they are easy to read, cite, and reuse.
Comparison of volatility forecasts from GARCH-family models (arch) and gradient boosting models (dquant) across 7 assets: BTC-USD, ETH-USD, EURUSD=X, BZ=F, GC=F, SI=F, SPY.
| Language | Web | Markdown |
|---|---|---|
| English | HTML | DQuant_Research.md |
| Russian | HTML | DQuant_Research_ru.md |
Research code: github.com/artrdon/DQuant_Research
Lab is meant to grow as an open collection of research related to DQuant, volatility forecasting, and quantitative trading.
If you have a paper, case study, benchmark, or methodology note you would like to publish here — contact the project author:
- Telegram: @Denchik_ai
- DQuant channel: @DQuant_Official
Please include a short summary, the type of material (research / applied / methodology), and links to code or data if available.
Research materials follow the licensing stated in each publication. The DQuant library is distributed under the MIT License.