How does the GAN based prediction in K. Zhang et al. (2018) improve performance?

Question

In Stock Market Prediction Based on Generative Adversarial Network by K. Zhang et. al, the authors feed financial data (X₀...X_t) into an LSTM to predict X_t+1.

Then, they evaluate whether the series (X₀...X_t+1) is real or not (with as X_t+1 either the predicted or the one appearing in the data).

This is done in a GAN system, with the predictor trained to create an X_t+1 so that the series would fool the discriminator.

What is the benefit of training the predictor as a generator in GAN, with the discriminator in providing the loss function, when it would be simpler to use a loss based on |predX_t+1 - X_t+1| ?

After all, the aim is to predict the target values - so why not train using the fit of the prediction directly?