I finished this project, i got a little bit confused when i was analyzing the errors values but after hours of reading and searching i finally got it (or not?)
Like always any feedback, corrections, everything it’s appreciated
Thank you and good morning, day, night wherever you are
Predicting The Stock Market.ipynb (73.5 KB)
Click here to view the jupyter notebook file in a new tab
Amazing work @alegiraldo666
There is this section ( Calculating average columns) where you’ve repeated your self severally It would be better if you could create create a function to do the same.
I am quite unsure how function works in this point because we are calculating mean() for different periods of data? Kindly advise if i am missing something here.
Btw @alegiraldo666 the project is cool and simple
Thanks for sharing with community.
Yeah, i think i’m going to work on it again this weekend and change a couple of things now that i’m finally not confused anymore on overfit and underfit
@prasadkalyan05 well i’ve been thinking and i think that function could be something like this
def average_time(col, w, s):
mean = stocks[col].window(w).mean().shift(s)
same with the std ones. Then just use .apply() and it’s done
Any of you guys have any thoughts on the errors values? I mean it’s clearly overfitted but at the same time as i said in my short analysis i guess it’s correct because to predict the value of the stock we actually need more data…
Anyway thank you both for the feedback!
Good to know on functions point.
As per my knowledge either we should remove some features or we need more data to train the model in order to overcome with over fitting.