corona/preprocess_data.R at master · nicebread/corona · GitHub

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
today <- Sys.Date()
recent_ECDC_file <- list.files("data/ECDC_data/", pattern="ECDC") %>% tail(1)
recent_CSSE_confirmed_file <- list.files("data/CSSE_data/", pattern="CSSE_confirmed") %>% tail(1)
recent_CSSE_deaths_file <- list.files("data/CSSE_data/", pattern="CSSE_deaths") %>% tail(1)
recent_CSSE_recovered_file <- list.files("data/CSSE_data/", pattern="CSSE_recovered") %>% tail(1)


# load data files
dat0_ECDC <- import(paste0("data/ECDC_data/", recent_ECDC_file))
dat0_CSSE_confirmed <- import(paste0("data/CSSE_data/", recent_CSSE_confirmed_file))
dat0_CSSE_deaths <- import(paste0("data/CSSE_data/", recent_CSSE_deaths_file))
dat0_CSSE_recovered <- import(paste0("data/CSSE_data/", recent_CSSE_recovered_file))

# load country level population data: How many people live in each country?
pop <- import("data/population_data/API_SP.POP.TOTL_DS2_en_csv_v2_821007.csv", header=TRUE)
pop$country <- pop[, 1]
pop$population <- pop[, "2018"]
pop <- pop %>% select(country, population) %>% arrange(country)
pop[pop$country == "Korea, Rep.", "country"] <- "South Korea"
pop[pop$country == "United States", "country"] <- "USA"
pop[pop$country == "Russian Federation", "country"] <- "Russia"
pop[pop$country == "Iran, Islamic Rep.", "country"] <- "Iran"
#print(sort(unique(pop$country)))

# load in US state population data
# State population data from 2019 US Census: https://www.census.gov/www/datasets/time-series/demo/popest/2010s-state-total.html
state_pop = read.csv('data/population_data/state_population_estimates_2019.csv', stringsAsFactors = FALSE) %>%
  mutate(., population = as.numeric(gsub(',', '', population_2019))) %>%
  dplyr::select(., -population_2019)


# ---------------------------------------------------------------------
#  preprocess the data sets, make common format

dat_ECDC <- dat0_ECDC %>%
	rename(
		# assign shorted var names
		country = countriesAndTerritories,
		new_cases = cases,
		new_deaths = deaths
	) %>%
	mutate(
		date = as.Date(paste0(year, "-", month, "-", day))
	)


dat_ECDC$country[dat_ECDC$country == "South_Korea"] <- "South Korea"
dat_ECDC$country[dat_ECDC$country == "United_States_of_America"] <- "USA"
dat_ECDC$country[dat_ECDC$country == "United_Kingdom"] <- "United Kingdom"


dat_ECDC <- dat_ECDC %>%
	group_by(country) %>%
	arrange(country, date) %>%
	mutate(
		cum_cases = cumsum(new_cases),
		cum_deaths = cumsum(new_deaths),
		overall_cum_cases = max(cum_cases),
		cum_cases_l1 = lag(cum_cases),
		dailyGrowth = cum_cases / cum_cases_l1 - 1,
		day_in_dataset = 1:n(),
		# country label only at the last data point of each timeline:
		country_label = if_else(day_in_dataset == max(day_in_dataset), as.character(country), NA_character_)
	)

dat_ECDC$dailyGrowth[is.nan(dat_ECDC$dailyGrowth) | is.infinite(dat_ECDC$dailyGrowth)] <- NA


dat_ECDC <- inner_join(dat_ECDC, pop, by="country")
dat_ECDC <- dat_ECDC %>%
  dplyr::mutate(
		cum_cases_per_100000 = cum_cases / (population/100000),
		cum_deaths_per_100000 = cum_deaths / (population/100000),
		cum_deaths_noZero = removeZero(cum_deaths),
		cum_deaths_per_100000_noZero = removeZero(cum_deaths_per_100000)
	)

ECDC_data_date <- max(dat_ECDC$date)
save(dat_ECDC, ECDC_data_date, file="data/dat_ECDC.RData")

# ---------------------------------------------------------------------
#  Preprocess CSSE data


# ---------------------------------------------------------------------
#  merge three CSSE data sets

# transform to long format for country data
dat_CSSE_confirmed <- dat0_CSSE_confirmed %>% select(-3, -4) %>% pivot_longer(-c(1:2), names_to="date.original", values_to="cum_cases")
dat_CSSE_deaths <- dat0_CSSE_deaths %>% select(-3, -4) %>% pivot_longer(-c(1:2), names_to="date.original", values_to="cum_deaths")
dat_CSSE_recovered <- dat0_CSSE_recovered %>% select(-3, -4) %>% pivot_longer(-c(1:2), names_to="date.original", values_to="cum_recovered")

dat_CSSE_combined <- inner_join(dat_CSSE_confirmed, dat_CSSE_deaths) # %>% inner_join(dat_CSSE_recovered)
colnames(dat_CSSE_combined)[2] <- c("country")

dat_CSSE_combined$country[dat_CSSE_combined$country == "Korea, South"] <- "South Korea"
dat_CSSE_combined$country[dat_CSSE_combined$country == "US"] <- "USA"
dat_CSSE_combined$country[dat_CSSE_combined$country == "Taiwan*"] <- "Taiwan"

# This data set is a mess ...
# Why is there a province "Recovered" in Canada?
# dat_CSSE_combined %>% pull(country) %>% unique()
# dat_CSSE_combined %>% pull(country) %>% table()
# dat_CSSE_combined %>% pull(country) %>% table() %>% table()
# dat_CSSE_combined %>% filter(country=="USA") %>% print(n=200)
# dat_CSSE_combined %>% filter(country=="Canada") %>% print(n=1000)
# dat_CSSE_combined %>% filter(country=="Germany") %>% print(n=1000)
# dat_CSSE_combined %>% filter(`Province/State`=="Recovered") %>% print(n=1000)


dat_CSSE0 <- dat_CSSE_combined %>%
	filter(
		!`Province/State` %in% c("Diamond Princess", "Recovered")
	) %>%
	group_by(country, date.original) %>%
	# aggregate countries which have multiple states in the data base
  summarise(
		cum_cases = sum(cum_cases),
		cum_deaths = sum(cum_deaths)#,
		#cum_recovered = sum(cum_recovered)
	) %>%
  ungroup() %>%
	mutate(
		date = mdy(date.original)
	) %>%
	group_by(country) %>%
	arrange(country, date) %>%
	mutate(
		overall_cum_cases = max(cum_cases),
		cum_cases_l1 = lag(cum_cases),
		dailyGrowth = cum_cases / cum_cases_l1 - 1,
		day_in_dataset = 1:n(),
		# country label only at the last data point of each timeline:
		country_label = if_else(day_in_dataset == max(day_in_dataset), as.character(country), NA_character_)
	)

dat_CSSE0$dailyGrowth[is.nan(dat_CSSE0$dailyGrowth) | is.infinite(dat_CSSE0$dailyGrowth)] <- NA


dat_CSSE <- inner_join(dat_CSSE0, pop, by="country") %>%
	mutate(
		cum_cases_per_100000 = cum_cases / (population/100000),
		cum_deaths_per_100000 = cum_deaths / (population/100000),
		cum_deaths_noZero = removeZero(cum_deaths),
		cum_deaths_per_100000_noZero = removeZero(cum_deaths_per_100000)
	)

CSSE_data_date <- max(dat_CSSE$date)

save(dat_CSSE, CSSE_data_date, file="data/dat_CSSE.RData")


# # US states: transform to long format with state-by-state US data
# dat_CSSE_US_states <- dat_CSSE_combined %>%
#   dplyr::select(., state = `Province/State`, everything()) %>%
#   dplyr::filter(., country == 'US', !grepl(',',state), state != 'Grand Princess') %>% #filter out county-level data
#   dplyr::select(-country) %>%
#   #pivot_longer(-1, names_to="date.original", values_to="cum_cases") %>%
#   #  group_by(state, date.original) %>%
#   #summarise(cum_cases = sum(cum_cases)) %>%
#   ungroup() %>%
#   mutate(
#     date = mdy(date.original)
#   ) %>%
#   group_by(state) %>%
#   arrange(state, date) %>%
#   mutate(
#     overall_cum_cases = max(cum_cases),
#     cum_cases_l1 = lag(cum_cases),
#     dailyGrowth = cum_cases / cum_cases_l1 - 1,
#     day_in_dataset = 1:n(),
#     # state label only at the last data point of each timeline:
#     state_label = if_else(day_in_dataset == max(day_in_dataset), as.character(state), NA_character_),
#     # keep country label as USA for all US state data
#     country = 'USA',
#     country_label = if_else(day_in_dataset == max(day_in_dataset), as.character(country), NA_character_)
#   ) %>%
#   left_join(state_pop, by = 'state') %>%
#   mutate(
# 		cum_cases_per_100000 = cum_cases / (population/100000),
# 		cum_deaths_per_100000 = cum_deaths / (population/100000),
		# cum_deaths_noZero = removeZero(cum_deaths),
		# cum_deaths_per_100000_noZero = removeZero(cum_deaths_per_100000)
# 	)
#
# dat_CSSE_US_states$dailyGrowth[is.nan(dat_CSSE_US_states$dailyGrowth) | is.infinite(dat_CSSE_US_states$dailyGrowth)] <- NA