DataTaunew | comments | leaders | submitlogin
Bitcoin price forecasting with LSTM and RNN models (
18 points by ibobriakov 9 days ago | 2 comments

2 points by blackbear 7 days ago | link

this is garbage. the partial autocorrelation clearly shows that successive values are highly correlated, and the model was trained to predict the price at time t using the price at time t-1: the model is basically learning to add a constant to its only input value. just look at the predictions on the test set


1 point by legel 4 days ago | link

Predicting markets with machine learning is always dangerous, because unlike most types of statistics, markets do not evidently have a single representation that is slowly uncovered; "gravity can flip upside down overnight".


RSS | Announcements