Open rowcolumn opened 5 years ago
The claim is the false positive predictions might still yield positive earnings (hopefully they would be greater than 0% but less than 10%). His next chart shows the data does not support this. The stock movement for the FPs is skewed toward negative returns.
Firstly, I appreciate your effort for using GAN to time series prediction.
In your write up, I find the number of false positive is large. How come a false prediction can make money as claimed in the write up?
Thank you.