Global Machine Learning Models Are the Future of Time Series Forecasting

After the M4 competition, conductors of the whole competition kind of concluded that all of these new H5NC models, like machine learning and deep learning, are not the best thing for type series forecasting. And this N-Bates model was able to perform better. If N-B Bates would have been part of the N-4 competition, they would have won that competition. The book that I was putting forward is also that you kind of start looking beyond eremas and profits and exponential smoothing and see that there is a whole different class of models, which works extremely well. It's just more suited to current situations. So we need to go ahead and kind of adopt