Last answered:

29 Jul 2020

Posted on:

27 Jul 2020

0

How to predict values TimeSeries?

Dear Team, Could you please help me with the following question? I have seen all the course but i am still not able to Predict with future data, I meant, I have the model created with auto_arima and SARIMAX, but when i try to insert future data it doesn't show me anything. Maybe i am not doing it in a good way, could you please send me an example how to predict future data when the data contains minutes? Like this: 2020-07-27 00:00:00+01:00 And as end date: 2020-07-27 17:20:00+01:00
KeyError: 'The `start` argument could not be matched to a location related to the index of the data.'
Thank you for your help  
1 answers ( 0 marked as helpful)
Instructor
Posted on:

29 Jul 2020

0
Hey, Miguel!   I think the issue here is the frequency of the data. If you're using data points of up to minutes (instead of days like we do in the course), you need to specify that when setting the frequency.      Here is a list of some of the most commonly used frequencies: https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#timeseries-offset-aliases   Additionally, when you're importing your data, are you making sure it's converting to a datetime type which includes hours and minutes, rather than stops at days? Here is a link to the to_datetime functions, which helps us convert imported dataset entries (usually saved as strings) into dates.  https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.to_datetime.html     Try to set the frequency to minutes and let me know if the issue is resolved. Otherwise, please provide some snippets of your, so I can have a better idea of where the code crashes.    Best, 365 Vik

Submit an answer