Time series dataset Predictions

Hi, I read some of the articles in Analytics Vidhya, I have a dataset that contains columns of time and 5 to 6 more columns and from these total of 6-7 columns, I have to predict 3 columns. How can I do that?

From my learning, I know multivariate time series, VAR can be used to predict multiple columns from a single time column, and AR is used to predict univariate time series problems, how should I get the above problem implemented?

I have done the time series prediction for 25 variables relating SDG data set for India. I made the forecast for each variable seperately.

I also want to know if there is a method for forecasting for all the 25 variables in one go.
I am trying to write a function for repeated forecast codes. There could be other methods.

Multivariate time series are series with multiple variables measured at each timestamp
Time | A | B | C
t – 3 | 0 | –1 | –2
t – 2 | 3 | –2 |–3
t – 1 | 4 | –2 |–4
t | 8 | –3 | –9

with Multivariate your aim is to predict at time t+1 what is value of A,B ,C
Models you can try VAR, Neural network

© Copyright 2013-2021 Analytics Vidhya