Hey! I love your videos. You kinda saved my ass for my thesis. I understood everything in your playlist for time series except the algorithms for finding the right parameter combination which you have mentioned in this video again. Thank you for making these videos!
Depends! If you are making y stationary, then you probably want x to be as well (if it isn't already) so it can be the best predictor. For example, If you make y stationary because it is trending, but x is also trending, then the trending x will probably no longer predict the now stationary y. Hope this helps!
I can't thank him enough for making this series and making it soo much fun
You should make more of these. A brilliantly lucid explanation
Hey!
I love your videos. You kinda saved my ass for my thesis. I understood everything in your playlist for time series except the algorithms for finding the right parameter combination which you have mentioned in this video again. Thank you for making these videos!
@3:23 a lot of bad memory from my calculus days 🤣🤣🤣🤣.
dealing with multivariate data, should all data be stationary or only y column data?
Depends! If you are making y stationary, then you probably want x to be as well (if it isn't already) so it can be the best predictor. For example, If you make y stationary because it is trending, but x is also trending, then the trending x will probably no longer predict the now stationary y. Hope this helps!
Your videos are very helpful. Sir, please make a video on the Functional Autoregressive model (FAR) model.
Do you know where I could find a good review of most popular/state of the art methods for timeseries forecasting?
The best thing would be Rob Hyndman's book found here:
otexts.com/fpp3/
@@AricLaBarr Thanks!
@@AricLaBarr The holy bible of time series analysis.)
Excellent. Could you please upload more? Thanks!
I love your videos. Can you make a video on Differential EMD? Thank you in adavance😊
Awesome !
Plz.Do more