应用截图

Tradingview assistant截图1
Tradingview assistant截图2
Tradingview assistant截图3
Tradingview assistant截图4

详细介绍

An assistant for backtesting trading strategies and showing external signals in Tradingview

Functionality
1. Backtesting trading strategies, optimisation of the strategy's parameters:
* automatic getting a list of parameters and their types (numeric, lists and checkboxes are supported)
* generation of the testing range according to the rule: the beginning value is 2 times less than the current one, the end is 2 times more than the current one.
* saving the generated parameters of testing a trading strategy for their correction as a template in a file in CSV format
* Loading adjusted parameter ranges from a CSV file
* Configuring the optimization model:
* Choosing the type of optimization: searching for the maximum or minimum values
* Selecting an optimized value from the entire list of strategy results in Tradingview (Net Profit, Ratio Avg Win / Avg Loss, Sharpe Ratio, Sortino Ratio, etc.)
* Choosing a search strategy in the parameter space(random, sequential, annealing method)
* Filtering of unsuitable results. For example, the number of tradings is less than necessary
* Setting the number of cycles to search for parameters.
* Performing automatic selection of parameters with storing all the results in the browser storage and the ability to save them as CSV files after testing, including in case of an error or page reloading
* Showing backtesting results on 3d chart to analyze the effect of various parameters on the result.

2. Upload external signals to Tradingview chart

Loading external buy or sell signals by timestamps from a CSV file

Optimization Methods

The sequential improvements optimization method is implement adjusting the best value already found. It does not perform a complete search of the entire parameter space.
The logic of it work is as follows. The current best state (parameters for max results) is taken. The first parameter is taken and all its values in the range are checked sequentially. If the best result is found, then further verificati