Indicators change their values in the range from -1 to 1. Bringing indicators to a common scale is useful for data analysis and machine learning. To normalize indicators, we use standardization using Z-Score. We also use vectors and standard vector methods to simplify the code.
Price_normaliz_1_0.mq5 - normalized price
entropy_normaliz_1_0.mq5 - normalized entropy
ATR_normaliz_1_0.mq5 - normalized ATR
CCI_normaliz_1_0.mq5 - normalized CCI
Pearson_price_index_1_0.mq5 - correlation of price with index without normalization, often used and also often mistakenly called autocorrelation
Pearson_price_price_1_0.mq5 - correlation of price with price offset by Lag bars, classic autocorrelation without normalization
Pearson_price_price_normaliz_1_0.mq5 - correlation of the price with the price shifted by Lag bars, classic autocorrelation with normalization
Similarly, you can normalize (standardize) any indicator on your own without much effort. If you remove tanh() from the indicators, values above 1 and below -1 may indicate anomalies.