Multiple linear regression from Statsmodels library coupled with Xarray library
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
examples
regressors
scripts
.gitignore
LICENSE.md
README.md
requirements.txt

README.md

X-regression

X-regression library should serve to regression of particular phenomena, e.g. 11-year solar cycle, and their climatic effects. Your feedback is highly appreciated.

Required package installation

pip install -r requirements.txt

Usage

Generate QBO orthogonal regressors via eofs_univ_ccmi_xarray.py script, e.g.:

python eofs_univ_ccmi_xarray.py -v jra55 u 1960 1960 2009 jra55_uwnd_1960_2009_zm.nc

When all regressors available, attribution by MLR to particular phehomena can be done via lin_reg_univ_ccmi_xarray.py script, e.g.:

python lin_reg_univ_ccmi_xarray.py -v jra55 t 1960 1979 2005 jra55_tmp_1960_2009_zm.nc all_trend

Possible acknowledgement via Zenodo:

DOI

References

Kuchar, A., Ball, W. T., Rozanov, E. V., Stenke, A., Revell, L., Miksovsky, J., Pisoft P., Peter, T. (2017). On the aliasing of the solar cycle in the lower stratospheric tropical temperature. Journal of Geophysical Research: Atmospheres. http://doi.org/10.1002/2017JD026948