Multivariate Timeseries Forecasting with few variables fixed for future

Course Q&A TensorFlow Developer Professional Sequences, Time Series and Prediction
1

When creating a post, please add:

  • Week # must be added in the tags option of the post.
  • Link to the classroom item you are referring to:
  • Description (include relevant info but please do not post solution code or your entire notebook):

Hi,

I am facing a challenge, where in a multivariate forecasting, I don’t have to forecast all the variables. There are 5 variables which are fixed for the future and are input by users, they impact my other variables which are supposed to be forecasted.

Could someone suggest a good resource which could help me solve this?

3

hi @Sahil_Mehta1

can you first confirm if you are talking about assignment from the course or your personal project. Also provide if all your variables dependent and independent variables are real time data?

Regards
DP

4

Hi @Deepti_Prasad,

Thanks for your reply!

This is a personal project. I will quickly explain the problem:

I have 10 variables as independent variables and 1 dependent variable.
Independent variables consist of categorical variables as well as continuous variables. Dependent variable is continuous time series data.

5

is the dependent variable real time series data or just a seasonal data?

You also need to explain what challenge are you facing. you have explain that part.

Kindy give a brief description of your dataset, what are you trying to achieve, what were the results and setback.