-
Notifications
You must be signed in to change notification settings - Fork 0
/
get_stats.R
294 lines (233 loc) · 15.5 KB
/
get_stats.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
codedir = "~/Desktop/github/COVID_LIC"
datadir = "~/paultangerusda drive/2020_Sync/COVID analysis (Paul Tanger)/data/"
setwd(codedir)
source('functions.R')
source('load_libs.R')
setwd(datadir)
subset.alls.plot = readRDS("subset.alls.plot_20200527_1159.RObj")
subset.alls.plot = readRDS("subset.alls.plot_20200616_1524.RDS")
compartments = c("cases", "death_o")
countries = levels(subset.alls.plot$USAID_Country)
AllAllsData.7scens.AgeAll.SelectCountries = subset.alls.plot[USAID_Country %in% countries & compartment %in% compartments & Scenarios=="Unmitigated"]
AllAllsData.7scens.AgeAll.SelectCountries = droplevels(AllAllsData.7scens.AgeAll.SelectCountries)
PeakMedAllCntryAllScensCases = as.data.table(AllAllsData.7scens.AgeAll.SelectCountries)
as.data.frame(colnames(PeakMedAllCntryAllScensCases))
PeakMedAllCntryAllScensCases = PeakMedAllCntryAllScensCases[,c(2,7,36,41,45,48,50)]
head(PeakMedAllCntryAllScensCases, 2)
# get quick stats on dates..
# so we want to get the JHU peak date..
# keep what we need
#PeakMedAllCntryAllScensCases = subset.alls.plot[,c("age","t","scen_id","Date","date_50"):=NULL]
# get peaks
PeakMedAllCntryAllScensCases2 = PeakMedAllCntryAllScensCases[
PeakMedAllCntryAllScensCases[, .I[which.max(med)], by=list(region_name, USAID_Country, compartment, Scenarios)]$V1]
# get peaks for UK
PeakMedAllCntryAllScensCases1 = PeakMedAllCntryAllScensCases[,-6]
PeakMedAllCntryAllScensCases2 = PeakMedAllCntryAllScensCases1[
PeakMedAllCntryAllScensCases1[, .I[which.max(med)], by=list(region_name, USAID_Country, compartment, Scenarios)]$V1]
# get peaks for JHU
PeakMedAllCntryAllScensCases2 = PeakMedAllCntryAllScensCases[,-7]
PeakMedAllCntryAllScensCases3 = PeakMedAllCntryAllScensCases2[
PeakMedAllCntryAllScensCases2[, .I[which.max(med)], by=list(region_name, USAID_Country, compartment, Scenarios)]$V1]
# drop med, we just want dates
PeakMedAllCntryAllScensCases2 = PeakMedAllCntryAllScensCases2[,-4]
# drop countries with no data
PeakMedAllCntryAllScensCases2 = PeakMedAllCntryAllScensCases2[!is.na(PeakMedAllCntryAllScensCases2$Date_JHU),]
# save this
setwd(datadir)
filename = addStampToFilename("peak_cases_deaths_all_countries", "csv")
write.csv(PeakMedAllCntryAllScensCases2, filename, row.names = F)
# can do it with dt functions?
# PeakMedAllCntryAllScensCases3 = PeakMedAllCntryAllScensCases2[, peak_diff := compartment.Date_JHU - compartment.Date_JHU, keyby=.(USAID_Country, region_abbrev, Scenarios)]
# PeakMedAllCntryAllScensCases3 = PeakMedAllCntryAllScensCases2[PeakMedAllCntryAllScensCases2, on=.(Date_JHU, compartment), peak_diff := Date_JHU - i.Date_JHU, by=.(USAID_Country, region_abbrev, Scenarios)]
# PeakMedAllCntryAllScensCases3 = PeakMedAllCntryAllScensCases2[, peak_diff := Date_JHU - Date_JHU, by=.(USAID_Country, region_abbrev, Scenarios)]
#
# data[data[condition=="dark"], on=.(measure,subject), v := score - i.score ]
# get range of peak death days
peak_death_dates = PeakMedAllCntryAllScensCases2[compartment == "death_o"]
deaths.summary = setDT(peak_death_dates)[, .(.N, Min=min(Date_UK),
Median=median(Date_UK),
Max=max(Date_UK)),
keyby=.(Scenarios)]
deaths.summary.region = setDT(peak_death_dates)[, .(.N, Min=min(Date_UK),
Median=median(Date_UK),
Max=max(Date_UK)),
keyby=.(Scenarios, region_name)]
# save these
setwd(datadir)
filename = addStampToFilename("peak_death_dates", "csv")
write.csv(deaths.summary, filename, row.names = F)
setwd(datadir)
filename = addStampToFilename("deaths.summary.region", "csv")
write.csv(deaths.summary.region, filename, row.names = F)
# make wide to calc day diff
PeaksbyCountry.wide = dcast.data.table(PeakMedAllCntryAllScensCases2, USAID_Country + region_name ~ Scenarios + compartment, value.var = "Date_UK")
# make it wide, but keep two cols so we can put into plots
PeaksbyCountry.wide.plots = dcast.data.table(PeakMedAllCntryAllScensCases2, USAID_Country + region_name + Scenarios ~ compartment, value.var = "Date_UK")
# renam cols to make it pretty
colnames(PeaksbyCountry.wide.plots) = c("USAID_Country", "Region", "Scenario", "Peak Cases", "Peak Deaths")
setwd(datadir)
filename = addStampToFilename("PeakCaseDeathDatesAllCountriesForPlots", "csv")
write.csv(PeaksbyCountry.wide.plots, filename, row.names = F)
# save this
setwd(datadir)
filename = addStampToFilename("PeakCaseDeathDatesAllCountries", "csv")
write.csv(PeaksbyCountry.wide, filename, row.names = F)
# get diffs the long way..
PeaksbyCountry.wide$unmit_diff = PeaksbyCountry.wide$Unmitigated_death_o - PeaksbyCountry.wide$Unmitigated_cases
PeaksbyCountry.wide$dist20_diff = PeaksbyCountry.wide$`20% distancing_death_o` - PeaksbyCountry.wide$`20% distancing_cases`
PeaksbyCountry.wide$dist50_diff = PeaksbyCountry.wide$`50% distancing_death_o` - PeaksbyCountry.wide$`50% distancing_cases`
PeaksbyCountry.wide$shield4080_diff = PeaksbyCountry.wide$`40/80 shielding_death_o` - PeaksbyCountry.wide$`40/80 shielding_cases`
PeaksbyCountry.wide$shield8080_diff = PeaksbyCountry.wide$`80/80 shielding_death_o` - PeaksbyCountry.wide$`80/80 shielding_cases`
PeaksbyCountry.wide$shield80dist20_diff = PeaksbyCountry.wide$`80/80 shielding, 20% distancing_death_o` - PeaksbyCountry.wide$`80/80 shielding, 20% distancing_cases`
PeaksbyCountry.wide$shield80dist50_diff = PeaksbyCountry.wide$`80/80 shielding, 50% distancing_death_o` - PeaksbyCountry.wide$`80/80 shielding, 50% distancing_cases`
# just keep diffs
as.data.frame(colnames(PeaksbyCountry.wide))
PeaksbyCountry.wide = PeaksbyCountry.wide[,c(1,2,17:23)]
head(PeaksbyCountry.wide)
# make long
PeaksbyCountry.long = melt(PeaksbyCountry.wide, id = c("USAID_Country", "region_abbrev"), variable.name = "Scenarios", value.name = "days_between_peaks")
PeaksbyCountryRegion.summary = setDT(PeaksbyCountry.long)[, .(Min=min(days_between_peaks),
Median=median(days_between_peaks),
Max=max(days_between_peaks)),
keyby=.(Scenarios, region_abbrev, USAID_Country)]
PeaksbyRegion.summary = setDT(PeaksbyCountry.long)[, .(Min=min(days_between_peaks),
Median=median(days_between_peaks),
Max=max(days_between_peaks)),
keyby=.(Scenarios, region_abbrev)]
Peaks.summary = setDT(PeaksbyCountry.long)[, .(Min=min(days_between_peaks),
Median=median(days_between_peaks),
Max=max(days_between_peaks)),
keyby=.(Scenarios)]
Peaks.summary$diff = Peaks.summary$Max - Peaks.summary$Min
PeaksbyCountry.range = Peaks.summary[,c(1,2,4)]
# save these
#PeaksbyCountryRegion.summary
filename = addStampToFilename("PeaksbyCountryRegion.summary", "csv")
write.csv(PeaksbyCountryRegion.summary, filename, row.names = F)
#PeaksbyRegion.summary
filename = addStampToFilename("PeaksbyRegion.summary", "csv")
write.csv(PeaksbyRegion.summary, filename, row.names = F)
#Peaks.summary
filename = addStampToFilename("Peaks.summary", "csv")
write.csv(Peaks.summary, filename, row.names = F)
#PeaksbyCountry.range
filename = addStampToFilename("PeaksbyCountry.range", "csv")
write.csv(PeaksbyCountry.range, filename, row.names = F)
####### ####### ####### ####### ####### ####### ####### ####### ####### #######
####### get cumulative cases for 12 months
# can do this from accs files, or the final tables from their sync folders
# but I'll do it from alls file since I already have that
# numbers may be slightly diff - flag for Katie
AllAllsData.7scens.AgeAll.SelectCountries = subset.alls.plot[USAID_Country %in% countries & compartment %in% compartments]
AllAllsData.7scens.AgeAll.SelectCountries = droplevels(AllAllsData.7scens.AgeAll.SelectCountries)
AllCntryAllScensCasesDeaths = as.data.table(AllAllsData.7scens.AgeAll.SelectCountries)
as.data.frame(colnames(AllCntryAllScensCasesDeaths))
CumulAllCntryAllScensCasesDeaths = AllCntryAllScensCasesDeaths[,c(2,7,36,41,45)]
head(CumulAllCntryAllScensCasesDeaths, 2)
SumsCumulAllCntryAllScensCasesDeaths = setDT(CumulAllCntryAllScensCasesDeaths)[, .(Cumul_Value=sum(med)), keyby=.(region_name, USAID_Country, Scenarios, compartment)]
# SumsCumulAllCntryAllScensCasesDeaths = CumulAllCntryAllScensCasesDeaths[CumulAllCntryAllScensCasesDeaths[, sum(med), by=list(USAID_Country, Scenarios, compartment)]$V1]
# nicer col names
colnames(SumsCumulAllCntryAllScensCasesDeaths) = c("Region", "Country", "Scenario", "Parameter", "Cumululative 1 yr total")
SumsCumulAllCntryAllScensCasesDeaths$`Cumululative 1 yr total` = round(SumsCumulAllCntryAllScensCasesDeaths$`Cumululative 1 yr total`)
SumsCumulAllCntryAllScensCasesDeaths
setwd(datadir)
filename = addStampToFilename("CumulAllCntryAllScensCasesDeaths", "csv")
write.csv(SumsCumulAllCntryAllScensCasesDeaths, filename, row.names = F)
####### ####### ####### ####### ####### ####### ####### ####### ####### ####### #######
####### ####### ####### ####### ####### ####### ####### ####### ####### ####### #######
####### ####### ####### ####### ####### ####### ####### ####### ####### ####### #######
# load the file above
setwd(datadir)
# get min and max too
MedPeakMedAllCntryAllScensCases = PeakMedAllCntryAllScensCases[
PeakMedAllCntryAllScensCases[, .I[which.max(med)], by=list(Country_OU, compartment,Scenarios)]$V1]
LoPeakMedAllCntryAllScensCases = PeakMedAllCntryAllScensCases[
PeakMedAllCntryAllScensCases[, .I[which.max(lo)], by=list(Country_OU, compartment,Scenarios)]$V1]
HiPeakMedAllCntryAllScensCases = PeakMedAllCntryAllScensCases[
PeakMedAllCntryAllScensCases[, .I[which.max(hi)], by=list(Country_OU, compartment,Scenarios)]$V1]
# put them together
Combo = cbind(LoPeakMedAllCntryAllScensCases, MedPeakMedAllCntryAllScensCases[,15], HiPeakMedAllCntryAllScensCases[,15])
# rename cols
names(Combo)[15] = 'JHU_Lo_Peak'
names(Combo)[16] = 'JHU_Med_Peak'
names(Combo)[17] = 'JHU_Hi_Peak'
# reassign to continue with below steps
PeakMedAllCntryAllScensCases2 = Combo
# TODO: could we do it together? not working yet
# MedPeaks <- PeakMedAllCntryAllScensCases[PeakMedAllCntryAllScensCases[, lapply(.SD, which.max()), .SDcols = 'med', by=list(Country_OU, compartment,Scenarios)]$V1]
# LoPeaks <- DT[, lapply(.SD, max), .SDcols = VarToMax, by = c(GrpVar1, GrpVar2)]
# HiPeaks <- DT[, lapply(.SD, max), .SDcols = VarToMax, by = c(GrpVar1, GrpVar2)]
# merge(DTmaxs, DTaves)
# drop NA
summary(PeakMedAllCntryAllScensCases2)
PeakMedAllCntryAllScensCases2 = na.omit(PeakMedAllCntryAllScensCases2, cols="Date_JHU")
PeakMedAllCntryAllScensCases2 = na.omit(PeakMedAllCntryAllScensCases2, cols="JHU_Med_Peak")
# save this in case want to use pivot tables
filename = addStampToFilename("PeakMedAllCntryAllScensCases2", "csv")
write.csv(PeakMedAllCntryAllScensCases2, filename, row.names = F)
filename = addStampToFilename("PeakMedAllCntryAllScensCases2LoHi", "csv")
write.csv(PeakMedAllCntryAllScensCases2, filename, row.names = F)
# load the file above
setwd(datadir)
PeakMedAllCntryAllScensCases2 = read.csv("official_output/PeakMedAllCntryAllScensCases2_20200517_1714.csv")
# set asdates
PeakMedAllCntryAllScensCases2$Date_JHU = as.Date(PeakMedAllCntryAllScensCases2$Date_JHU)
# instead of just cases below, just get the ranges of everything
PeakMedAllCntryAllScens.Ranges.n.etc = PeakMedAllCntryAllScensCases2[, .( Min=min(Date_JHU, na.rm=T), Median=median(Date_JHU), Max=max(Date_JHU, na.rm=T)), by=c('Continent', 'USAID_region', 'compartment', 'Scenarios')]
# but this won't work because need to get min of lo and hi dates separately
PeakMedAllCntryAllScens.Ranges.n.etc = setDT(PeakMedAllCntryAllScensCases2)[, .(.N, Min=min(Date_JHU, na.rm=T), Median=median(Date_JHU), Max=max(Date_JHU, na.rm=T)), by=c('Continent', 'USAID_region', 'compartment', 'Scenarios')]
# with combo var above
LoHiMedPeakMedAllCntryAllScens.Ranges.n.Combo = setDT(PeakMedAllCntryAllScensCases2)[, .(.N, Min=min(JHU_Lo_Peak, na.rm=T), Median=median(JHU_Med_Peak), Max=max(JHU_Hi_Peak, na.rm=T)), by=c('Continent', 'USAID_region', 'compartment', 'Scenarios')]
# just med
PeakMedAllCntryAllScens.Ranges.n.etc = setDT(PeakMedAllCntryAllScensCases2)[, .(.N, Median=median(Date_JHU)), by=c('Continent', 'USAID_region', 'compartment', 'Scenarios')]
# export this
setwd(datadir)
filename = addStampToFilename("DateStats_AllCountriesBins", "csv")
write.csv(PeakMedAllCntryAllScens.Ranges.n.etc, filename, row.names = F)
# include the subregions and remove non USAID presence countries
PeakMedAllCntryAllScensCases3 = PeakMedAllCntryAllScensCases2[USAID_presence != 'N']
PeakMedAllCntryAllScens.Ranges.n.etc = setDT(PeakMedAllCntryAllScensCases3)[, .(.N, Median=median(Date_JHU)), by=c('Continent', 'USAID_region', 'USAID_subregion', 'compartment', 'Scenarios')]
# save this
setwd(datadir)
filename = addStampToFilename("DateStats_AllCountriesBinsWithSubRegionsAndPresence", "csv")
write.csv(PeakMedAllCntryAllScens.Ranges.n.etc, filename, row.names = F)
# TODO: include list of countries in the output (instead of just N)
# TODO: get min and max
PeakMedAllCntryAllScens.min = setDT(PeakMedAllCntryAllScensCases3)[, .(.N, Min=min(Date_JHU)), by=c('Continent', 'USAID_region', 'USAID_subregion', 'compartment', 'Scenarios')]
PeakMedAllCntryAllScens.max = setDT(PeakMedAllCntryAllScensCases3)[, .(.N, Max=max(Date_JHU)), by=c('Continent', 'USAID_region', 'USAID_subregion', 'compartment', 'Scenarios')]
# just cases
PeakMedAllCntryAllScensCases2 = PeakMedAllCntryAllScensCases2[compartment == "cases"]
# so we want the range of dates for each scenario (for all countries)
# maybe later do for USAID countries
summary(PeakMedAllCntryAllScensCases2$Date_JHU)
with(PeakMedAllCntryAllScensCases2, tapply(Date_JHU, Scenarios, range))
filename = addStampToFilename("DateStats_AllCountries", "txt")
setwd(datadir)
sink(filename)
with(PeakMedAllCntryAllScensCases2, tapply(Date_JHU, Scenarios, range))
sink()
# what is death rate?
# load file
subset.alls.plot = readRDS("subset.alls.plot_20200527_1159.RObj")
compartments = c("cases", "death_o")
countries = levels(subset.alls.plot$LSHTM_Country)
AllAllsData.7scens.AgeAll.SelectCountries = subset.alls.plot[LSHTM_Country %in% countries & compartment %in% compartments]
AllAllsData.7scens.AgeAll.SelectCountries = droplevels(AllAllsData.7scens.AgeAll.SelectCountries)
countrieslist = split(AllAllsData.7scens.AgeAll.SelectCountries, by="USAID_Country")
AllAllsData.7scens.AgeAll.SelectCountries$USAID_Country = as.factor(AllAllsData.7scens.AgeAll.SelectCountries$USAID_Country)
countries = names(countrieslist)
death.dt = as.data.table(AllAllsData.7scens.AgeAll.SelectCountries)
death.dt = death.dt[,c(2,33,38,42)]
# dcast
death.dt.wide = dcast(death.dt, USAID_Country + Scenarios ~ compartment, value.var = "med", fun.aggregate = sum)
death.dt.wide$death_rate = (death.dt.wide$death_o / death.dt.wide$cases) * 100
# death.dt.wide = as.data.table(death.dt.wide)
# calculate
# death_rate = death.dt.wide[, as.list(summary(death.dt.wide$death_rate)), by=c('USAID_Country', 'Scenarios')]
# death_rate <- aggregate(.~Scenarios, death.dt.wide, FUN=c(min, median, max))
death.dt.wide = death.dt.wide[,c(1,2,5)]
death_rate2 <- aggregate(death_rate ~ Scenarios, data=death.dt.wide, FUN=function(x) c(min(x), median(x), max(x)))
death_rate2 <- cbind(death_rate2[,1], as.data.frame(death_rate2[,2]))
names(death_rate2) <- c("Scenario", "min", "median", "max")
# get diff between peak case and peak death day for each country and then get range