How to predict Stock Prices with Python using Facebook's prediction tool fbprophet

This video is containing an application of the prediction tool developed by the Data Science team of facebook - fbprophet.
You can apply this tool to time series data such as sales. I am applying it do a Stock Price Prediction here.
CREDITS TO FACEBOOK AND THE TOOL.
If you have problems getting the modules installed or imported please drop me a comment. It's a bit of a challenge but once you made it you have a very nice and powerful tool.

Facebooks Github:

Alternative installation on conda:

On Prompt install Ephem:
conda install -c anaconda ephem

Install Pystan:
conda install -c conda-forge pystan

Finally install Fbprophet
conda install -c conda-forge fbprophet

CREDITS TO LUANAFORMIGA on GitHub

Fixing holiday bug error:

You need to find the fbprophet folder on your pc first by a simple folder search for 'fbprophet'

from holidays import WEEKEND, HolidayBase, easter, rd

to

from holidays import WEEKEND, HolidayBase

CREDITS: ARTURGONTIJO on Github!

[sanjaydhande]

Thanks for such nice video series. The series is very helpful, to the point. Great explanation with good examples. Thanks again!

[hijamgyaneswarsingh]

Thank you for this video and it really helped me to understand the concept of facebook prophet. Sir, I'm facing some issues even after installation in different environment python version 3.8 but it failed to import in jupyter notebook. No module found Prophet

[saumyadeepsur]

Thanks a lot for this. It was quite helpful. Just a small question though- How to put legends in the final Forecast graph?

[Британскиеучёные-чх]

great video! Thank you very much! Listen, I don't understand English, can you answer one question? Why are the GOOG shares on the chart so different from your model? I mean, why doesn't your model take into account the data we know? Why are the black dots and the blue line diverging?

[xwclsg]

Thanks for this great video. I was wondering whether we could get the predicted distribution by using Prophet. More specifically, can we get the predicted conditional quantiles of the predictand? Many thanks in advance.

[modchat]

Thank you for making me rich :-). Very good explanation good job! Schöne Grüße aus Hamburg

[shahedahmed]

Thank you so much man!!! btw How do I calculate RMSE here?

[Think-tn]

Why does "exception model has not been fit" appear when running can I solve it？

[tlghnkck]

Hello, when I apply this software to indices (SP500, XU100), the graphic below is broken. What would be the reason. The same error occurs in parities such as (EURUSD, GBPUSD).

[jaiprathapgv]

Sir, I have another doubt. Why we are getting error when we don't add two columns ds and y?

[jaiprathapgv]

Sir, How should I predict the value of the stock on a particular date?

[elu]

installation changed: conda install -c conda-forge prophet

[imtim]

Hello, why can’t I just simply type “pip install fbprophet” on Jupyter notebook? I could not install neither Pystan nor fbprophet, plz help

[sigmaakhil]

Sir...I got error in installing FbProphet... please help

[mohammadsalvatore]

hello
I have problem with model.fit(df)

this is the problem :
INFO:fbprophet:Disabling daily seasonality. Run prophet with daily_seasonality=True to override this.

KeyError Traceback (most recent call last)
KeyError: 'metric_file'

then the kernal restarts