2
$\begingroup$

I'm currently working on a project that analyzes property values over time. I'm having a conceptual roadblock. The data are formatted as individual house prices for individual units over time, leaving me with thousands of individual time series to forecast. Is there an analysis that would allow me to forecast home prices for individual units over time? Are there any resources readily available in python? Moreover, is there an analysis that would allow me to add external regressors (e.g., local school quality)?

$\endgroup$
1
$\begingroup$

Conceptually you are looking for hierarchical times series forecasting. There is a package readily available for doing that in R. I don't know about any similar Python packages.

$\endgroup$
0
$\begingroup$

I think you should try VAR models with LASSO variable selection. I found a Python library for time series, including VAR models (but I didn't see LASSO): http://www.statsmodels.org/dev/tsa.html

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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