Zack Argyle | Engineering manager, Core Experience
The idea of building a “Progressive Web App” (PWA) is not new, but its definition has changed with the emergence of key technologies like service workers. Now it’s finally possible to build great…
Igor, you have a bug in this line:
prediction2_inverse = np.array(prediction_inverse[:,0][1:])
When you fix it, all your errors will increase dramatically, as they should, because you can’t forecast asset prices this way.
You are doing prediction wrong. Why do you think your validation plot has your prediction lagging your true value by one? (the prediction plot sits below the actual plot).
You model is not learning, it’s simply waiting for the next time step and output what the true value is, ie a lagged version of your test data.