To save me having to use separate summary functions, I need to include calculation of summary information in extract_mc_results.
I also need to be careful and more clever with the calculation of the summaries, in particular getting the distances calculated appropriately.
These are really slow at the moment as I do an epic joining approach, whereas I could probably write some nicer code in c++ to handle the summary statistics.
A few extra notes
# Some sort of summary table that returns the following columns
# n_added (n_added)
# This is taken from mc_model$n_added
# **n_added_location (for each urban/rural)**
# This is taken from group_by(mc_model,first(n_added))
# **location (is it urban/rural/overall)**
# This is taken from group_by(mc_model) + ungroup()
# **n_ohca (number OHCA events that occured)**
# This is taken from nrow(mc_model$user)
# n_ohca_cov (number of OHCA events that are covered)
# nrow(mc_model$users_affected) + sum(dat_dist$is_covered)
# total_aeds (total number of AEDs (original AEDs + n_added))
# nrow(mc_model$facility) + nrow(mc_model$facilities_added)
# pct_ohca_cov
# (nrow(mc_model$users_affected) + sum(dat_dist$is_covered))
# /
# nrow(mc_model$users)
# mean_dist_to_aed
# round(mean(dat_dist$distance),2)
# sd_dist_to_aed
# round(sd(dat_dist$distance),2)
njtierney
commented
7 years ago
To save me having to use separate summary functions, I need to include calculation of summary information in
extract_mc_results.I also need to be careful and more clever with the calculation of the summaries, in particular getting the distances calculated appropriately.
These are really slow at the moment as I do an epic joining approach, whereas I could probably write some nicer code in c++ to handle the summary statistics.
A few extra notes