0
$\begingroup$

I have two data sets one is cross sectional census data with 10 years interval and another one is time series data (monthly) for several years. Now I want to perform statistical time series analysis on this both dataset in a single model. Is it possible?

$\endgroup$

1 Answer 1

0
$\begingroup$

It really depends on the outcome you want from your analysis.

The most straightforward approaches might be resampling:

  • Interpolate the 10-year census data to month granularity. The interpolation might be linear, polynomial, etc. This effectively imposes a strong assumption in the dynamics of the data. This may make sense for some variables but not for others. And you are probably introducing a lot of variance in the obtained conclusions.

  • Aggregate the monthly time series into 10-year granularity. This may not be very useful if you want to predict something with an horizon less than 10 years...

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.