Backtesting trading strategies with NodeRed and MachineTrade

Backtesting trading strategies with NodeRed and MachineTrade udemy course free download

Backtesting and optimizing trading algorithms using historical time series data.

This course provides a detailed walkthrough on using the Machine Trader back tester to test and refine trading algorithms using historical data. The back tester enables users to simulate algorithm performance and make necessary adjustments for improved trading results.

My firm, MachineTrader, has developed a customized Node-Red platform for use in trading stocks, cryptos, and options through which you can create, manage, and execute your automated trading strategies. To do automated trading, however, you will need a cloud server where all of the prewritten scripts for communicating with a broker/dealer are stored as well as any trading scripts you create. Each MachineTrader™ trading subscription includes a hosted website which can expand to your unique requirements. This also means that all of your data and strategies are completely private - your algorithms remain proprietary to you!

Students will be given access to their own proprietary, password-protected trading platform by signing up for a MachineTrader Learner's Account under the terms of a 30-day free trial. After 30-days, the Learner’s Account renews at $24.95 (US) per month which should give you plenty of time to complete the course and decide if you want to continue using MachineTrader.

The tutorial begins with downloading prewritten code from a shared drive and importing it into a Node-RED workflow. The process involves setting up a database table to store price data, using Polygon as the data source, and ensuring that historical stock prices are properly fetched and recorded. The example focuses on Apple’s stock, though the process can be applied to any ticker symbol by modifying a single variable.

The next step involves structuring the table to store minute-by-minute trading data, creating 390 rows corresponding to each trading minute in a day. The data is then retrieved from Polygon via an HTTP request, storing price points in an array.

The trading engine is then engaged to simulate trades based on price data using a Bollinger Bands-based Z-score strategy. Trades occur when the Z-score crosses set thresholds, with the algorithm buying when the stock is oversold and selling when it is overbought. The tutorial explores optimizing the strategy by adjusting the Z-score threshold, testing multiple scenarios, and comparing profitability.

Finally, the data is exported to Excel for analysis, where trade performance, price movement, and Z-score behavior are visualized using charts. The course emphasizes the importance of data analysis in algorithmic trading, highlighting the iterative nature of backtesting for refining profitable strategies.