Martin Lysy, Hind A. Al-Abadleh, Lucas Neil, Priyesh Patel, Wisam Mohammed, Yara Khalaf
The paper can be found here.
Please see the package vignettes: vignette("aq2020") and vignette("aq2020-pvalue").
Please see the following scripts in inst/extdata:
aq2020-pollutant_extdata.R and aq2020-pollutant_pvalue.R: Preprocessing and p-value calculations for the pollutant data.
aq2020-temperature_extdata.R and aq2020-temperature_pvalue.R: Preprocessing and p-value calculations for the temperature data.
aq2020-irradiance_extdata.R and aq2020-irradiance_pvalue.R: Preprocessing and p-value calculations for the solar irradiance data.
aq2020-pval_plots.R: P-value boxplots in the paper.
Turn scripts into vignettes.
Standardize capitalization of columns in datasets.
Replace poll_* by pollutant_*.
Document pollutant datasets.
Document temperature and irradiance datasets. Should also clean up temperature_info.
Add 3-day webscrape function get_aqo3d(). What’s missing is to format the output in H00:H23 format (always EST).
Use get_aqo3d() to fill in November-December 2020 for Downtown_Toronto, after the station changed locations. Should store the new Station_ID though.
Filter pollutant_info on has_poll.
Perhaps pollutant_info and temperature_info don’t need to be stored in data, but rather only in inst/ext_data. That is, could easily recover the relevant information from pollutant_data and temperature_data, especially since we want to add station_id as identifier now, since station can have more than one of these.