博客
关于算法交易、回测策略、量化分析和数据质量的文章、指南和见解。
1–9 of 113 articles
ENMy Secret Is Theft: Why the Best Algorithmic Traders Steal Ideas, Not Invent Them
A viral confession that 'my secret is theft' captures an uncomfortable truth about trading system development: originality is overrated, and disciplined adaptation of known ideas usually beats invention.
ENTransaction Costs for Quants: Test 3 Regimes, Watch Engine Risk
Learn how quants can model commissions, spreads, slippage, and market impact, stress test three cost regimes, and account for engine and data risks.
ENSix Bar Aggregation Methods Quants Can Calibrate for ML Backtests
Compare six bar aggregation methods for backtesting and machine learning, then calibrate thresholds and validate results with clean intraday data.
ENSpread Modeling for Backtests: 3 Estimators to Stress Test ML Edge
Compare Roll, Corwin Schultz, and CHL spread estimates; model slippage and impact, then validate ML backtests with walk forward tests and clean minute bars.
ENMT4 History Center: 5 Safe Fixes Before File Repair
Fix MT4 History Center errors with safe refresh and download steps, .hst file repair, and clean data imports for more reliable, reproducible backtests.
ENWhen 1 Minute vs 5 Minute Bars Mislead Quants, Run a Signature Plot
Decide whether 1 minute or 5 minute bars suit your backtest. Follow a research-backed checklist: signature plot, walk forward validation, realistic...
ENMT4 Max Bars for Quants: Fix Broken Backtests with Clean Import Data
Understand how MT4’s chart and history limits work. Change Max bars, troubleshoot missing data, and import clean datasets to produce reproducible backtests.
ENPrevent Broken Backtests: Intraday Data Sources Practitioners Must Audit
Intraday data source checklist for practitioners: verify timestamp provenance, bar labels, and spike detection. Test a sample first.
ENWhy MT5 EAs Are Trending Again: A Practical Look at the Renewed Interest in Expert Advisors
MT5 Expert Advisors are seeing a fresh wave of search interest. Here's what's driving it, what separates a well-tested EA from a risky one, and how to evaluate automated strategies before relying on them.