The Multiple Non-Linear Regression indicator performs regression analysis using four independent price variables: close, open, high, and low. It normalizes these inputs within a specified data length to a 0–1 range based on their highest and lowest values. The indicator calculates a weighted regression equation with coefficients that are iteratively adjusted via gradient descent to minimize prediction errors. Users can customize parameters including normalization length, learning rate, initial coefficients, and smoothing options.

Visualization features include plotting normalized input data points as circular markers, displaying the regression line with a color gradient reflecting its value and bar colors, and presenting normalized data alongside predicted values in a table. The indicator also signals crossovers of the regression line with a midline at 0.5, which can be interpreted as potential buy or sell signals. This tool supports analysis across various markets and symbols such as BTCUSD, Forex, Crypto, XAUUSD, and NAS100, aiding users in understanding relationships between multiple price variables to inform trading decisions.