I am trying to implement neural networks - LSTMs - to implement a Time Series concept. I have 3 years of hourly data of production and I plan to use 2 years of data to train the model and predict for the third year of data to test the model.

I want to make a prediction(target variable - Production) for each hour, given that I have 1 independent variable which varies monthly(say the 'Price'), another which varies hourly(say the 'Demand') and a couple of other variables which vary with different categories present in my data. I also have a couple of other continuous value variables which do not necessarily vary with time, there is no fixed pattern with respect to time.

Can I use time series for such data? If yes, what are the ways to treat the monthly varying quantity('Price'), the hourly varying quantity('Demand'), and time-invariant quantities?

I also have a variable which holds numerical values (say Production capacity of a Unit) that change which a categorical variable (say the Unit name) in a dataset. Do I treat the "Production Capacity" - as a continuous value variable or a category?

Below is a small section of how my data looks like: enter image description here

New contributor
shweta24 is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.

Your Answer

shweta24 is a new contributor. Be nice, and check out our Code of Conduct.

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.