Min-Max Rescaling

Min-max rescaling is a normalization technique that shifts and rescales data to a fixed range, typically between zero and one. This is achieved by subtracting the minimum value from the data point and dividing by the range, which is the difference between the maximum and minimum values.

It is highly effective for preparing data for neural networks and other machine learning models where feature scaling is a prerequisite. In the domain of derivatives, it allows for the comparison of diverse indicators like RSI, volume, and funding rates on a uniform scale.

By constraining the data, it prevents features with large magnitudes from dominating the learning process. This method preserves the relative relationships between data points, ensuring that the structural integrity of the signal is maintained.

It is a simple yet powerful tool for feature engineering in predictive modeling. When applied to financial time series, it helps in visualizing multiple indicators on a single chart without overlap.

It remains one of the most accessible ways to standardize input data for quantitative analysis.