Here is a perfectly replicated TradingView backtesting engine condensed into a single library function calculated with arrays. It includes TradingView's calculations for Net profit, Total Trades, Percent of Trades Profitable, Profit Factor, Max Drawdown (absolute and percent), and Average Trade (absolute and percent). Here's how TradingView defines each aspect of its backtesting system: Net Profit: The overall profit or loss achieved. Total Trades: The total number of closed trades, winning and losing. Percent Profitable: The percentage of winning trades, the number of winning trades divided by the total number of closed trades. Profit Factor: The amount of money the strategy made for every unit of money it lost, gross profits divided by gross losses. Max Drawdown: The greatest loss drawdown, i.e., the greatest possible loss the strategy had compared to its highest profits. Average Trade: The sum of money gained or lost by the average trade, Net Profit divided by the overall number of closed trades.
Here's how each variable is defined in the library function: _backtest(bool _enter, bool _exit, float _startQty, float _tradeQty)
bool _enter: When the strategy should enter a trade (entry condition)
bool _exit: When the strategy should exit a trade (exit condition)
float _startQty: The starting capital in the account (for BTCUSD, it is the amount of USD the account starts with)
float _tradeQty: The amount of capital traded (if set to 1000 on BTCUSD, it will trade 1000 USD on each trade)
Currently, this library only works with long strategies, and I've included a commented out section under DEMO STRATEGY where you can replicate my results with TradingView's backtesting engine. There's tons I could do with this beyond what is shown, but this was a project I worked on back in June of 2022 before getting burned out. Feel free to comment with any suggestions or bugs, and I'll try to add or fix them all soon. Here's my list of thing to add to the library currently (may not all be added):
Add commission calculations.
Add support for shorting
Add a graph that resembles TradingView's overview graph.
Clean and optimize code.
Clean up in a way that makes it easy to add other TradingView calculations (such as Sharpe and Sortino ratio).
Separate all variables, so they become accessible outside of calculations (such as gross profit, gross loss, number of winning trades, number of losing trades, etc.).
진정한 TradingView 정신에 따라, 저자는 이 파인 코드를 다른 파인 프로그래머들이 재사용할 수 있도록 오픈 소스 라이브러리로 공개했습니다. 저자에게 박수를 보냅니다! 이 라이브러리는 개인적으로 사용하거나 다른 오픈 소스 출판물에서 사용할 수 있지만, 이 코드를 출판물에서 재사용하는 것은 하우스 룰에 의해 관리됩니다.