Skip to content

Noncrossing quantile smoothing splines for temperature/dew point analysis

Notifications You must be signed in to change notification settings

karenamckinnon/humidity_variability

Repository files navigation

humidity_variability

This repo contains code to fit the noncrossing smoothing splines quantile regression model in McKinnon and Poppick (JABES, https://link.springer.com/article/10.1007/s13253-020-00393-4)

Code is research (not production) quality, and will likely need modifications to run on your machine.

  • ./humidity_variability/scripts/save_gsod.py: Get GSOD data used in analysis
  • ./humidity_variability/scripts/create_cases.py: Create synthetic data and fit model to the synthetic case studies
  • ./humidity_variability/main.py: Fit either the fully interaction model or a simpler linear model to GSOD data. This function is designed to run across multiple processors if available.

Note that the interaction model takes 2-6 seconds to run for each quantile, so fitting a set of 19 quantiles (5th to 95th percentile in steps of 5%) will take about a minute.

Please contact Karen McKinnon ([email protected]) if you are using the code or methods in your work.

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

About

Noncrossing quantile smoothing splines for temperature/dew point analysis

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages