Split-Adjusting Your Dividend History
Split-adjusting dividend history: why a 4-for-1 split fakes a fourfold yield spike, and the two repairs that fix the per share payout series. Worked on AAPL.
#backtesting
Filtering by topic #backtesting · clear
How Much Slippage to Assume in a Backtest
How much slippage does a backtest need to assume? Half the quoted spread is a floor. Here is why that floor is optimistic and how to measure the real number.
What Breaks an Options Backtest
What breaks an options backtest is the entry price and the contract list. Reprice every entry one volatility point and see how much modelled credit survives.
Event-Driven vs Vectorized Backtesting
Event-driven vs vectorized backtesting, settled with data: what the same-bar fill costs in index points, and the three things only an event loop can model.
Open-Source Trading Terminal for Python
What an open-source trading terminal for Python really is: where an agentic IDE sits between a notebook, a backtester and a live broker, and what it cannot do.
January Barometer: Does It Actually Work?
Does the January Barometer work? We score January's sign against the base rate of positive years over full history, and again with January excluded.
When Historical Market Data Gets Revised
Historical market data gets revised: how a daylight saving bug dropped after-hours bars, the four revision types, and the record that keeps results checkable.
Backtesting in Illiquid Markets: Fill Models
Backtesting in illiquid markets: how three fill models change the same strategy's reported return, and the checks that tell an artifact from an edge.
Do Volume Indicators Predict Anything?
Do volume indicators predict anything? We code OBV and VPT as published, define a falsifiable divergence rule, and measure forward returns against a baseline.
Is the Market a Random Walk? Real Returns
Is the market a random walk? Twenty years of daily SPY returns, measured: the statistics that cannot tell real from simulated, and the two that always can.
Latency Models in HFT Backtests Explained
Latency models in HFT backtests decide which fills are real. See how feed delay and order entry delay differ, and what a zero latency fill quietly hides.
The Open-Source TradingView Optimizer
The open-source TradingView optimizer grid searches strategy inputs and ranks the winner. Here is the overstatement it never prints, worked on real data.
How to Backtest a Trading Strategy in Python
How to backtest a trading strategy in under 80 lines of plain Python: a 5/20 moving average crossover with next-open fills, fees, slippage and a split fix.
What Is Backtesting in Trading?
Backtesting replays a trading rule over past bars, seeing only what was known at the time. What comes out, four ways it lies, and when a result is too good.