📄️ List of independent variables used
The selection of independent variables was painstaking and reworked progressively during testing. The initial, non-exhaustive list included around twenty technical indicators, but the final version contains exactly seven.
📄️ Python Replication
Exportation was done from both APIs bitcoin-data.com and complemented with fallback data from Blockchain.com info are on a daily basis, and I computed daily data to get weekly ones.
📄️ Dependent Variable
The dependent variable must be normalized to be viable for use and to train the neural network model. In this case, after experimenting with several types of normalization that were not sufficiently precise, I decided to normalize my independent variable on a scale ranging from -1 to 1, where -1 corresponds to the lowest level of a cycle and 1 corresponds to the highest level of a cycle.