-
Notifications
You must be signed in to change notification settings - Fork 1
/
evaluation_code_4.R
514 lines (424 loc) · 27.8 KB
/
evaluation_code_4.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
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
#Conduct mpox nowcasting evaluation
#Analyst: Rebecca Rohrer
#Code review: Allegra Wilson
#Last updated 8/12/2024
## Run this code file fourth (4/4) ##
#This code generates figures that visualize the comparison of by-day and by-week hindcasts of reported and estimated
#but not-yet-reported number of confirmed and probable cases among New York City residents
#diagnosed with mpox from July 8 through September 27, 2022.
# This code populates:
# Figure 2
# Figure S1
# Figure S2
# Figure S3
#By diagnosis date
## Figure 2 (daily) ----
#For the dx dataset to plot we need:
#1) Daily, the actual case count
dateseq_daily <- seq(as.Date("2022-07-20"), as.Date("2022-09-27"), by = 1) %>% #7/20 is the first date an estimate is produced for 14-day for a nowcast conducted on a Tuesday
as.data.frame()
dsd_dx <- dateseq_daily
names(dsd_dx) <- "dx_date"
dsd_dx$dx_date <- as.Date(dsd_dx$dx_date)
plot_totals <- adf %>%
filter(dx_date <= as.Date("2022-09-27")) %>%
group_by(dx_date) %>%
dplyr::summarize(estimate = n()) %>%
full_join(dsd_dx, by = "dx_date") %>%
mutate(method= "Actual")
plot_totals <- replace(plot_totals, is.na(plot_totals), 0)
#2) Daily, the daily case count from each nowcast output (ex. 14-day, dx nowcast)
#just the method, onset_var, and estimate columns
dx14 <- ManageNowcastOutputToGraph("comparison_master_all_dx14.csv", "14 day", include_weekend = TRUE, type="daily")
dx21 <- ManageNowcastOutputToGraph("comparison_master_all_dx21.csv", "21 day", include_weekend = TRUE, type="daily")
dx28 <- ManageNowcastOutputToGraph("comparison_master_all_dx28.csv", "28 day", include_weekend = TRUE, type="daily")
dx35 <- ManageNowcastOutputToGraph("comparison_master_all_dx35.csv", "35 day", include_weekend = TRUE, type="daily")
dx42 <- ManageNowcastOutputToGraph("comparison_master_all_dx42.csv", "42 day", include_weekend = TRUE, type="daily")
dx49 <- ManageNowcastOutputToGraph("comparison_master_all_dx49.csv", "49 day", include_weekend = TRUE, type="daily")
#Combine, not stratified
combined_overall_dx <- plot_totals %>%
full_join(dx14, by=c("dx_date"="onset_var", "method", "estimate")) %>%
full_join(dx21, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
full_join(dx28, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
full_join(dx35, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
full_join(dx42, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
full_join(dx49, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
mutate(dx_date = as.Date(dx_date))
#Combine all data
data_toplot <- plot_totals %>%
full_join(combined_overall_dx, by=c("dx_date", "method", "estimate")) %>%
mutate(upper = as.numeric(upper),
lower = as.numeric(lower),
upper_edit = case_when((method == "14 day") ~ upper,
TRUE ~ as.numeric(NA)),
lower_edit = case_when((method == "14 day") ~ lower,
TRUE ~ as.numeric(NA)),
upper_edit_crop = case_when(upper_edit >= 120 ~ 120, #cutting off above 120 95% PI for clarity
TRUE ~ upper_edit),
method = factor(method, levels = c("Actual", "14 day", "21 day", "28 day", "35 day", "42 day", "49 day")),
dx_date = as.Date(dx_date))
WriteCSVToFolder(data_toplot, "dx_daily_data_for_fig2.csv")
#Make x axis labels at weekly, though there will be tick marks at daily
data_toplot_labels <- data_toplot %>%
mutate(label = ifelse(dx_date == "2022-07-08" | dx_date == "2022-09-27", format(dx_date, "%d%b"),
ifelse( lubridate::wday(dx_date) == 3,
format( dx_date, "%d%b" ),
"" ) ))
y_labels_daily <- c(0,"",20,"",40,"",60,"",80,"",100,"",120)
p_dx <- ggplot(data_toplot) +
geom_ribbon(data=data_toplot %>% filter(!is.na(lower_edit)), aes(x=dx_date, ymin=lower_edit, ymax=upper_edit_crop,
group=interaction(method),
fill=method), alpha=.15,
show.legend=FALSE) +
geom_line(aes(x=dx_date, y=estimate, group=interaction(method), color=method)) +
geom_point(aes(x=dx_date, y=estimate, group=interaction(method), color=method, shape=method)) +
scale_shape_manual(name="Nowcasting scenario", values =c("Actual" = 1,"14 day" =2,"21 day"=3,"28 day"=4,"35 day"=5,"42 day"=6,"49 day"=7)) +
scale_color_manual(name="Nowcasting scenario", values = c("#000000","#5a5a5a","#fc8d62","#8da0cb","#e78ac3","#a6d854","#ffd92f")) +
scale_fill_manual(values = c("#5a5a5a"), name="Nowcasting scenario") +
scale_x_date(breaks = data_toplot$dx_date, labels = data_toplot_labels$label, expand=c(0,0)) +
scale_y_continuous(expand=c(0,0), breaks=seq(0,125,10), labels=y_labels_daily)+
theme_classic() +
labs(x="Diagnosis date (daily)",
y = "Confirmed and probable cases",
color= "Nowcasting scenario") +
theme(axis.text.x = element_text(angle = 90, vjust = .5),
panel.grid.major.x = element_blank(),
panel.grid.minor.x = element_blank(),
strip.text = element_text(face = "bold")) +
theme(text = element_text(family = "Times New Roman"))
## Figure S1 (weekly) ----
#1) Weekly, not stratified the actual case count Tues-Weds
dx2 <- ManageNowcastOutputToGraph("comparison_master_all_dx2.csv", "2 week", include_weekend = TRUE, type="weekly")
dx3 <- ManageNowcastOutputToGraph("comparison_master_all_dx3.csv", "3 week", include_weekend = TRUE, type="weekly")
dx4 <- ManageNowcastOutputToGraph("comparison_master_all_dx4.csv", "4 week", include_weekend = TRUE, type="weekly")
dx5 <- ManageNowcastOutputToGraph("comparison_master_all_dx5.csv", "5 week", include_weekend = TRUE, type="weekly")
dx6 <- ManageNowcastOutputToGraph("comparison_master_all_dx6.csv", "6 week", include_weekend = TRUE, type="weekly")
dx7 <- ManageNowcastOutputToGraph("comparison_master_all_dx7.csv", "7 week", include_weekend = TRUE, type="weekly")
dateseq_weekly <- seq(as.Date("2022-07-26"), as.Date("2022-09-27"), by = 7) %>% #starting at first date 7/26 when an estimate is produced
as.data.frame()
dsw <- dateseq_weekly
names(dsw) <- "onset_var"
dsw$onset_var <- as.Date(dsw$onset_var)
adf_prep <- adf %>% #starting at first date 7/26 when an estimate is produced
filter(dx_date <= as.Date("2022-09-27")) %>%
mutate(onset_var = case_when("2022-07-08" <= dx_date & dx_date <= "2022-07-12" ~ "2022-07-12", #not a full week, week 0 - should be none of these
"2022-07-13" <= dx_date & dx_date <= "2022-07-19" ~ "2022-07-19",
"2022-07-20" <= dx_date & dx_date <= "2022-07-26" ~ "2022-07-26",
"2022-07-27" <= dx_date & dx_date <= "2022-08-02" ~ "2022-08-02",
"2022-08-03" <= dx_date & dx_date <= "2022-08-09" ~ "2022-08-09",
"2022-08-10" <= dx_date & dx_date <= "2022-08-16" ~ "2022-08-16",
"2022-08-17" <= dx_date & dx_date <= "2022-08-23" ~ "2022-08-23",
"2022-08-24" <= dx_date & dx_date <= "2022-08-30" ~ "2022-08-30",
"2022-08-31" <= dx_date & dx_date <= "2022-09-06" ~ "2022-09-06",
"2022-09-07" <= dx_date & dx_date <= "2022-09-13" ~ "2022-09-13",
"2022-09-14" <= dx_date & dx_date <= "2022-09-20" ~ "2022-09-20",
"2022-09-21" <= dx_date & dx_date <= "2022-09-27" ~ "2022-09-27")
) %>% #should be none of these
mutate(onset_var = as.Date(onset_var))
plot_totals_weekly <- adf_prep %>%
filter(onset_var <= as.Date("2022-09-27")) %>% #end date of actual counts should also stop at final point 9/27
group_by(onset_var) %>%
summarise(estimate=n()) %>% #calling actual varname estimate so it can merge with nowcast outputs
full_join(dsw, by = c("onset_var")) %>%
mutate(method = "Actual")
plot_totals_weekly <- replace(plot_totals_weekly, is.na(plot_totals_weekly), 0)
#Combined, not stratified
combined_overall_dx_weekly <- plot_totals_weekly %>%
full_join(dx2, by=c("onset_var", "method", "estimate")) %>%
full_join(dx3, by=c("onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(dx4, by=c("onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(dx5, by=c("onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(dx6, by=c("onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(dx7, by=c("onset_var", "method", "estimate", "upper", "lower")) %>%
mutate(dx_date = as.Date(onset_var))
#Combine all data
data_toplot_weekly <- plot_totals_weekly %>%
full_join(combined_overall_dx_weekly, by=c("onset_var", "method", "estimate")) %>%
mutate(method = factor(method, levels = c("Actual", "2 week", "3 week", "4 week", "5 week", "6 week", "7 week")),
onset_var = as.Date(onset_var),
upper = as.numeric(upper),
lower = as.numeric(lower),
upper_edit = case_when((method == "2 week") ~ upper,
TRUE ~ as.numeric(NA)),
lower_edit = case_when((method == "2 week") ~ lower,
TRUE ~ as.numeric(NA)),
upper_edit_crop = case_when(upper_edit >= 600 ~ 600, #crop overall at 600
TRUE ~ upper_edit)) %>%
filter(onset_var > "2022-07-12") #remove first incomplete week
WriteCSVToFolder(data_toplot_weekly, "dx_weekly_data_for_figS1.csv")
p_dx_weekly <- ggplot(data_toplot_weekly) +
geom_ribbon(data=data_toplot_weekly %>% filter(!is.na(lower_edit)), aes(x=onset_var, ymin=lower_edit, ymax=upper_edit_crop,
group=interaction(method),
fill=method), alpha=.15,
show.legend=FALSE) +
geom_line(aes(x=onset_var, y=estimate, group=interaction(method), color=method)) +
geom_point(aes(x=onset_var, y=estimate, group=interaction(method), color=method, shape=method)) +
scale_shape_manual(name = "Nowcasting scenario", values =c("Actual"=1,"2 week"=2,"3 week"=3,"4 week"=4,"5 week"=5,"6 week"=6,"7 week"=7)) +
scale_color_manual(values = c("#000000","#5a5a5a","#fc8d62","#8da0cb","#e78ac3","#a6d854","#ffd92f")) +
scale_fill_manual(values = c("#5a5a5a"), name="Nowcasting scenario") +
scale_x_date(date_labels = "%b-%d", breaks = data_toplot_weekly$onset_var, expand=c(0,0)) +
scale_y_continuous(expand=c(0,0))+
theme_classic() +
labs(x="Diagnosis week ending",
y = "Confirmed and probable cases",
color= "Nowcasting scenario") +
theme(axis.text.x = element_text(angle = 90, vjust = .5),
panel.grid.major.x = element_blank(),
panel.grid.minor.x = element_blank(),
strip.text = element_text(face = "bold")) +
theme(text = element_text(family = "Times New Roman"))
#By illness onset date, non-stratified
##Figure S2 (daily) ----
#For the onset dataset to plot we need:
#1) the actual case count
dsd_onset <- dateseq_daily
names(dsd_onset) <- "onset_date"
dsd_onset$onset_date <- as.Date(dsd_onset$onset_date)
plot_totals <- adf_onset %>%
filter(onset_date <= as.Date("2022-09-27")) %>%
group_by(onset_date) %>%
dplyr::summarize(estimate = n()) %>%
full_join(dsd_onset, by = "onset_date") %>%
mutate(method= "Actual")
plot_totals <- replace(plot_totals, is.na(plot_totals), 0)
#2) the daily case count from each nowcast output (ex. 14-day, dx nowcast)
#just the method, onset_var, and n columns
onset14 <- ManageNowcastOutputToGraph("comparison_master_all_onset14.csv", "14 day", include_weekend = TRUE, type="daily")
onset21 <- ManageNowcastOutputToGraph("comparison_master_all_onset21.csv", "21 day", include_weekend = TRUE, type="daily")
onset28 <- ManageNowcastOutputToGraph("comparison_master_all_onset28.csv", "28 day", include_weekend = TRUE, type="daily")
onset35 <- ManageNowcastOutputToGraph("comparison_master_all_onset35.csv", "35 day", include_weekend = TRUE, type="daily")
onset42 <- ManageNowcastOutputToGraph("comparison_master_all_onset42.csv", "42 day", include_weekend = TRUE, type="daily")
onset49 <- ManageNowcastOutputToGraph("comparison_master_all_onset49.csv", "49 day", include_weekend = TRUE, type="daily")
combined_overall_onset <- plot_totals %>%
full_join(onset14, by=c("onset_date"="onset_var", "method", "estimate")) %>%
full_join(onset21, by=c("onset_date"="onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(onset28, by=c("onset_date"="onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(onset35, by=c("onset_date"="onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(onset42, by=c("onset_date"="onset_var", "method", "estimate", "upper", "lower")) %>%
full_join(onset49, by=c("onset_date"="onset_var", "method", "estimate", "upper", "lower")) %>%
mutate(onset_date = as.Date(onset_date))
data_toplot <- plot_totals %>%
full_join(combined_overall_onset, by=c("onset_date", "method", "estimate")) %>%
mutate(upper = as.numeric(upper),
lower = as.numeric(lower),
upper_edit = case_when((method == "21 day") ~ upper,
TRUE ~ as.numeric(NA)),
lower_edit = case_when((method == "21 day") ~ lower,
TRUE ~ as.numeric(NA)),
upper_edit_crop = case_when(upper_edit >= 120 ~ 120,
TRUE ~ upper_edit),
method = factor(method, levels = c("Actual", "14 day", "21 day", "28 day", "35 day", "42 day", "49 day")),
onset_date = as.Date(onset_date))
#Make x axis labels at weekly, though there will be tick marks at daily
data_toplot_labels <- data_toplot %>%
mutate( label = ifelse(onset_date == "2022-07-08" | onset_date == "2022-09-27", format(onset_date, "%d%b"),
ifelse( lubridate::wday(onset_date) == 3,
format( onset_date, "%d%b" ),
"" ) ))
WriteCSVToFolder(data_toplot, "onset_daily_data_for_figS2.csv")
p_onset <- ggplot(data_toplot) +
geom_ribbon(data=data_toplot %>% filter(!is.na(lower_edit)), aes(x=onset_date, ymin=lower_edit, ymax=upper_edit_crop, fill=method), alpha=.15,
show.legend=FALSE) +
geom_line(aes(x=onset_date, y=estimate, color=method, group=interaction(method))) +
geom_point(aes(x=onset_date, y = estimate, group=interaction(method), color=method, shape=method)) +
scale_shape_manual(name="Nowcasting scenario", values =c("Actual" = 1,"14 day" =2,"21 day"=3,"28 day"=4,"35 day"=5,"42 day"=6,"49 day"=7)) +
scale_color_manual(values = c("#000000","#66c2a5","#5a5a5a","#8da0cb","#e78ac3","#a6d854","#ffd92f")) +
scale_fill_manual(values = c("#5a5a5a"), name="Nowcasting scenario") +
scale_x_date(breaks = data_toplot$onset_date, labels= data_toplot_labels$label, expand=c(0,0)) +
scale_y_continuous(expand=c(0,0), breaks=seq(0,125,10), labels=y_labels_daily) +
theme_classic() +
labs(x="Onset date (daily)",
y = "Confirmed and probable cases",
color= "Nowcasting scenario") +
theme(axis.text.x = element_text(angle = 90, vjust = .5),
panel.grid.major.x = element_blank(),
panel.grid.minor.x = element_blank(),
strip.text = element_text(face = "bold")) +
theme(text = element_text(family = "Times New Roman"))
##Figure S3 (weekly) ----
#1) Weekly, not stratified the actual case count Weds-Weds
#Wednesday dates to name week numbers "week ending" that Wednesday (ex. week 1 is the week ending 07-20-2022)
onset2 <- ManageNowcastOutputToGraph("comparison_master_all_onset2.csv", "2 week", include_weekend = TRUE, type="weekly")
onset3 <- ManageNowcastOutputToGraph("comparison_master_all_onset3.csv", "3 week", include_weekend = TRUE, type="weekly")
onset4 <- ManageNowcastOutputToGraph("comparison_master_all_onset4.csv", "4 week", include_weekend = TRUE, type="weekly")
onset5 <- ManageNowcastOutputToGraph("comparison_master_all_onset5.csv", "5 week", include_weekend = TRUE, type="weekly")
onset6 <- ManageNowcastOutputToGraph("comparison_master_all_onset6.csv", "6 week", include_weekend = TRUE, type="weekly")
onset7 <- ManageNowcastOutputToGraph("comparison_master_all_onset7.csv", "7 week", include_weekend = TRUE, type="weekly")
adf_onset_prep <- adf_onset %>%
filter(onset_date <= as.Date("2022-09-27")) %>%
mutate(onset_var = case_when("2022-07-08" <= onset_date & onset_date <= "2022-07-12" ~ "2022-07-12", #not a full week, week 0 - should be none of these
"2022-07-13" <= onset_date & onset_date <= "2022-07-19" ~ "2022-07-19",
"2022-07-20" <= onset_date & onset_date <= "2022-07-26" ~ "2022-07-26",
"2022-07-27" <= onset_date & onset_date <= "2022-08-02" ~ "2022-08-02",
"2022-08-03" <= onset_date & onset_date <= "2022-08-09" ~ "2022-08-09",
"2022-08-10" <= onset_date & onset_date <= "2022-08-16" ~ "2022-08-16",
"2022-08-17" <= onset_date & onset_date <= "2022-08-23" ~ "2022-08-23",
"2022-08-24" <= onset_date & onset_date <= "2022-08-30" ~ "2022-08-30",
"2022-08-31" <= onset_date & onset_date <= "2022-09-06" ~ "2022-09-06",
"2022-09-07" <= onset_date & onset_date <= "2022-09-13" ~ "2022-09-13",
"2022-09-14" <= onset_date & onset_date <= "2022-09-20" ~ "2022-09-20",
"2022-09-21" <= onset_date & onset_date <= "2022-09-27" ~ "2022-09-27")) %>%
mutate(onset_var = as.Date(onset_var))
plot_totals_weekly_onset <- adf_onset_prep %>%
filter(onset_var <= as.Date("2022-09-27")) %>%
group_by(onset_var) %>%
summarise(estimate=n()) %>% #calling actual varname estimate so it can merge with nowcast outputs
full_join(dsw, by = c("onset_var")) %>%
mutate(method = "Actual")
plot_totals_weekly_onset <- replace(plot_totals_weekly_onset, is.na(plot_totals_weekly_onset), 0)
#Combined, not stratified
combined_overall_onset_weekly <- plot_totals_weekly_onset %>%
full_join(onset2, by=c("onset_var", "method", "estimate")) %>%
full_join(onset3, by=c("onset_var", "method", "estimate", "lower", "upper")) %>%
full_join(onset4, by=c("onset_var", "method", "estimate", "lower", "upper")) %>%
full_join(onset5, by=c("onset_var", "method", "estimate", "lower", "upper")) %>%
full_join(onset6, by=c("onset_var", "method", "estimate", "lower", "upper")) %>%
full_join(onset7, by=c("onset_var", "method", "estimate", "lower", "upper")) %>%
mutate(onset_date = as.Date(onset_var))
#Combine all data
data_toplot_weekly_onset <- plot_totals_weekly_onset %>%
full_join(combined_overall_onset_weekly, by=c("onset_var", "method", "estimate")) %>%
mutate(method = factor(method, levels = c("Actual", "2 week", "3 week", "4 week", "5 week", "6 week", "7 week")),
onset_var = as.Date(onset_var),
upper = as.numeric(upper),
lower = as.numeric(lower),
upper_edit = case_when((method == "3 week") ~ upper,
TRUE ~ as.numeric(NA)),
lower_edit = case_when((method == "3 week") ~ lower,
TRUE ~ as.numeric(NA)),
upper_edit_crop = case_when(upper_edit >= 1200 ~ 1200, #crop overall at 1200
TRUE ~ upper_edit)) %>%
filter(onset_var > "2022-07-12") #remove first incomplete week
WriteCSVToFolder(data_toplot_weekly_onset, "onset_weekly_data_for_figS3.csv")
p_onset_weekly <- ggplot(data_toplot_weekly_onset) +
geom_ribbon(data=data_toplot_weekly_onset %>% filter(!is.na(lower_edit)), aes(x=onset_var, ymin=lower_edit, ymax=upper_edit_crop,
group=method,
fill=method), alpha=.15,
show.legend=FALSE) +
geom_line(aes(x=onset_var, y=estimate, group=interaction(method), color=method)) +
geom_point(aes(x=onset_var, y=estimate, group=interaction(method), color=method, shape=method)) +
scale_shape_manual(name = "Nowcasting scenario", values =c("Actual"=1,"2 week"=2,"3 week"=3,"4 week"=4,"5 week"=5,"6 week"=6,"7 week"=7)) +
scale_color_manual(values = c("#000000","#66c2a5","#5a5a5a","#8da0cb","#e78ac3","#a6d854","#ffd92f")) +
scale_fill_manual(values = c("#5a5a5a"), name="Nowcasting scenario") +
scale_x_date(date_labels = "%b-%d", breaks = data_toplot_weekly_onset$onset_var, expand=c(0,0)) +
scale_y_continuous(expand=c(0,0))+
theme_classic() +
labs(x="Onset week ending",
y = "Confirmed and probable cases",
color= "Nowcasting scenario") +
theme(axis.text.x = element_text(angle = 90, vjust = .5),
panel.grid.major.x = element_blank(),
panel.grid.minor.x = element_blank(),
strip.text = element_text(face = "bold")) +
theme(text = element_text(family = "Times New Roman"))
#Look at & save each plot
p_dx
ggsave(filename="evaluation_figure_2.png", path=dataset_path, plot=p_dx, height = 8, width = 10, unit = "in")
p_dx_weekly
ggsave(filename="evaluation_figure_S1.png", path=dataset_path, plot=p_dx_weekly, height = 8, width = 10, unit = "in")
p_onset
ggsave(filename="evaluation_figure_S2.png", path=dataset_path, plot=p_onset, height = 8, width = 10, unit = "in")
p_onset_weekly
ggsave(filename="evaluation_figure_S3.png", path=dataset_path, plot=p_onset_weekly, height = 8, width = 10, unit = "in")
#Example of generating a stratified nowcasting evaluation plot
#This code assumes you have generated "comparison_master_all_dx14_re.csv", "comparison_master_all_dx21_re.csv", etc.
# files containing the estimate data for all windows
#Daily totals, not stratified
# combined_overall_dx <- plot_totals %>%
# full_join(dx14, by=c("dx_date"="onset_var", "method", "estimate")) %>%
# full_join(dx21, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
# full_join(dx28, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
# full_join(dx35, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
# full_join(dx42, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
# full_join(dx49, by=c("dx_date"="onset_var", "method", "upper", "lower", "estimate")) %>%
# mutate(dx_date = as.Date(dx_date)) %>%
# mutate(race_ethnicity = "Overall") #doing this so the facet plot picks up overall as a 4th facet
# #1) Daily, totals, stratified
# re_date_shell <- as.data.table(dsd_dx) %>%
# mutate(baa = "Black",
# hl = "Hispanic/Latino",
# w = "White") %>%
# tidyr::pivot_longer(cols=c(baa, hl, w), values_to = "race_ethnicity") %>%
# select(DIAGNOSIS_DATE, race_ethnicity)
#
#
# plot_totals_re <- adf %>%
# mutate(race_ethnicity = case_when(race_ethnicity == "Black/African American" ~ "Black",
# TRUE ~ race_ethnicity)) %>%
# filter(DIAGNOSIS_DATE <= as.Date("2022-09-27")) %>%
# group_by(DIAGNOSIS_DATE, race_ethnicity) %>%
# filter(race_ethnicity == "Black" |
# race_ethnicity == "Hispanic/Latino" |
# race_ethnicity == "White") %>%
# summarize(estimate = n()) %>%
# full_join(re_date_shell, by = c("DIAGNOSIS_DATE", "race_ethnicity")) %>%
# mutate(method= "Actual")
#
# plot_totals_re <- replace(plot_totals_re, is.na(plot_totals_re), 0)
#
# #2) Daily, the daily case count from each nowcast output, stratified (ex. 14-day, dx nowcast)
# dx14_re <- ManageNowcastOutputToGraph("comparison_master_all_dx14_re.csv", "14 day", include_weekend = TRUE, r_e = TRUE, type="daily")
#
# dx21_re <- ManageNowcastOutputToGraph("comparison_master_all_dx21_re.csv", "21 day", include_weekend = TRUE, r_e = TRUE, type="daily")
#
# dx28_re <- ManageNowcastOutputToGraph("comparison_master_all_dx28_re.csv", "28 day", include_weekend = TRUE, r_e = TRUE, type="daily")
#
# dx35_re <- ManageNowcastOutputToGraph("comparison_master_all_dx35_re.csv", "35 day", include_weekend = TRUE, r_e = TRUE, type="daily")
#
# dx42_re <- ManageNowcastOutputToGraph("comparison_master_all_dx42_re.csv", "42 day", include_weekend = TRUE, r_e = TRUE, type="daily")
#
# dx49_re <- ManageNowcastOutputToGraph("comparison_master_all_dx49_re.csv", "49 day", include_weekend = TRUE, r_e = TRUE, type="daily")
#
#
# #Combine all data
# data_toplot <- plot_totals_re %>%
# full_join(combined_overall_dx, by=c("DIAGNOSIS_DATE", "method", "race_ethnicity", "estimate")) %>%
# full_join(dx14_re, by=c("DIAGNOSIS_DATE"="onset_var", "method", "race_ethnicity", "estimate", "upper", "lower")) %>%
# full_join(dx21_re, by=c("DIAGNOSIS_DATE"="onset_var", "method", "race_ethnicity", "estimate", "upper", "lower")) %>%
# full_join(dx28_re, by=c("DIAGNOSIS_DATE"="onset_var", "method", "race_ethnicity", "estimate", "upper", "lower")) %>%
# full_join(dx35_re, by=c("DIAGNOSIS_DATE"="onset_var", "method", "race_ethnicity", "estimate", "upper", "lower")) %>%
# full_join(dx42_re, by=c("DIAGNOSIS_DATE"="onset_var", "method", "race_ethnicity", "estimate", "upper", "lower")) %>%
# full_join(dx49_re, by=c("DIAGNOSIS_DATE"="onset_var", "method", "race_ethnicity", "estimate", "upper", "lower")) %>%
# mutate(upper = as.numeric(upper),
# lower = as.numeric(lower),
# upper_edit = case_when((method == "14 day") ~ upper,
# TRUE ~ as.numeric(NA)),
# lower_edit = case_when((method == "14 day") ~ lower,
# TRUE ~ as.numeric(NA)),
# upper_edit_crop = case_when(upper_edit >= 120 ~ 120, #cutting off above 120 95% PI for clarity
# TRUE ~ upper_edit),
# method = factor(method, levels = c("Actual", "14 day", "21 day", "28 day", "35 day", "42 day", "49 day")),
# DIAGNOSIS_DATE = as.Date(DIAGNOSIS_DATE),
# race_ethnicity = factor(race_ethnicity, levels=c("Overall", "Black", "Hispanic/Latino", "White"))
# )
#
#
# #Make x axis labels at weekly, though there will be tick marks at daily
# data_toplot_labels <- data_toplot %>%
# mutate(label = ifelse(DIAGNOSIS_DATE == "2022-07-08" | DIAGNOSIS_DATE == "2022-09-27", format(DIAGNOSIS_DATE, "%d%b"),
# ifelse( lubridate::wday(DIAGNOSIS_DATE) == 3,
# format( DIAGNOSIS_DATE, "%d%b" ),
# "" ) ))
#
# p_dx <- ggplot(data_toplot) +
# geom_ribbon(data=data_toplot %>% filter(!is.na(lower_edit)), aes(x=DIAGNOSIS_DATE, ymin=lower_edit, ymax=upper_edit_crop,
# group=interaction(method, race_ethnicity),
# fill=method), alpha=.15,
# show.legend=FALSE) +
# geom_line(aes(x=DIAGNOSIS_DATE, y=estimate, group=interaction(method, race_ethnicity), color=method)) +
# geom_point(aes(x=DIAGNOSIS_DATE, y=estimate, group=interaction(method, race_ethnicity), color=method, shape=method)) +
# scale_shape_manual(name="Nowcasting scenario", values =c("Actual" = 1,"14 day" =2,"21 day"=3,"28 day"=4,"35 day"=5,"42 day"=6,"49 day"=7)) +
# scale_color_manual(name="Nowcasting scenario", values = c("#000000","#5a5a5a","#fc8d62","#8da0cb","#e78ac3","#a6d854","#ffd92f")) +
# scale_fill_manual(values = c("#5a5a5a"), name="Nowcasting scenario") +
# scale_x_date(breaks = data_toplot$DIAGNOSIS_DATE, labels = data_toplot_labels$label, expand=c(0,0)) +
# scale_y_continuous(expand=c(0,0), breaks=seq(0,125,10), labels=y_labels_daily)+
# theme_classic() +
# labs(x="Diagnosis date (daily)",
# y = "Confirmed and probable cases",
# color= "Nowcasting scenario") +
# theme(axis.text.x = element_text(angle = 90, vjust = .5),
# panel.grid.major.x = element_blank(),
# panel.grid.minor.x = element_blank(),
# strip.text = element_text(face = "bold")) +
# facet_rep_grid(race_ethnicity ~ ., scales="free_y", repeat.tick.labels="all") +
# theme(text = element_text(family = "Times New Roman"))