QUANT SPECIAL · DATA CLEANING · PART 3 OF 8
本月免费剩余 2/3
Ask ZChat · save an investment note
Data Cleaning Part 3: Look-Ahead Bias and Survivorship Bias in Backtesting
For every backtest there are two default assumptions. First, the backtest engine knew only what was knowable at the moment of execution. Second, the engine saw every liquid stock that was available to trade at that moment.

Further reading
- VOL RESEARCHWhat Is Data Cleaning in Trading?Data cleaning finds and fixes wrong, missing and mistimed price data before it reaches your backtest.
- VOL RESEARCHTRADING DATA CLEANING Part 2: OHLCV Data Explained for BacktestingOHLCV data records each bar's open, high, low, close and volume, and its timestamp decides what your backtest can know.
- VOL RESEARCHForced Sellers Move Markets More Than Headlines DoThe 2022 UK gilt crisis, the August 2024 VIX spike, and every quarterly index rebalance trace back to the same mechanism — someone running out of room to wait
- VOL RESEARCHWhy Most Quant Traders Are Optimizing the Wrong ThingHere's what most quant traders have done wrong...
- VOL RESEARCHQuant Trading Is Not What You ThinkQuant trading is REALLY not what you think...
- VOL RESEARCHDeep Quant Flow Series 02 Retail Flow: The Cheapest Signal in Modern MarketsDeep Quant Flow Series 02 Retail Flow: The Cheapest Signal in Modern Markets Why free trading was never free, and why your order is worth more than your opinion Most retail traders