LSTM (Long Short-Term Memory)
A recurrent neural network architecture designed to learn long-range dependencies in sequential data such as price history.
Long Short-Term Memory networks are a type of recurrent neural network (RNN) built to remember information over long sequences. Gating mechanisms let the network decide what to keep, forget, and output at each step, which solves the vanishing-gradient problem that plagues vanilla RNNs.
For stock forecasting, LSTMs are well suited to capturing momentum and mean-reversion patterns that play out over weeks. endeavr.ai's LSTM variant adds an attention layer so the model can weight the most relevant historical days when forming a prediction.