-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathFigure 4.R
140 lines (118 loc) · 6.26 KB
/
Figure 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
setwd("~/EpiEstim/Epi Methods/Code Final")
source("Model_synthetic structure.R")
##################Choose parameter values and simulate model
rho=0.1
k=2
N=35039
20785/35039
14254/35039
Ni=floor(35039*0.6)
Nj=floor(35039*0.4)
Ei0=20
Ej0=20
Ii0=18
Ij0=18
Ri0=0
Rj0=0
one <- data.frame(run=seq(1,6,1),dist=rep(1,6),betai=c(4*10^-5,rep(4*10^-5,5)),
betaj=c(4*10^-5,rep(6*10^-5,5)),
betaij=c(4*10^-5,1*10^-5,1*10^-6,1*10^-7,1*10^-8,0))
two <- data.frame(run=seq(1,6,1),dist=rep(2,6),betai=c(4*10^-5,rep(4*10^-5,5)),
betaj=c(4*10^-5,rep(8*10^-5,5)),
betaij=c(4*10^-5,1*10^-5,1*10^-6,1*10^-7,1*10^-8,0))
three <- data.frame(run=seq(1,6,1),dist=rep(3,6),betai=c(4*10^-5,rep(4*10^-5,5)),
betaj=c(4*10^-5,rep(10*10^-5,5)),
betaij=c(4*10^-5,1*10^-5,1*10^-6,1*10^-7,1*10^-8,0))
four <- data.frame(run=seq(1,6,1),dist=rep(4,6),betai=c(4*10^-5,rep(4*10^-5,5)),
betaj=c(4*10^-5,rep(12*10^-5,5)),
betaij=c(4*10^-5,1*10^-5,1*10^-6,1*10^-7,1*10^-8,0))
pars <- rbind(one, two, three, four)
pars$R0 <- rep(NA,nrow(pars))
nsim=100
Rt_threshold_list <- list()
for (i in 1:nrow(pars)){
pars$R0[i] <- R_t_func(Xi=Ni-Ei0-Ii0-Ri0, Xj=Nj-Ej0-Ij0-Rj0,
betai=pars$betai[i],betaj=pars$betaj[i],betaij=pars$betaij[i],
gamma=gamma)
make_data %>%
simulate(
params=c(betai=pars$betai[i],betaj=pars$betaj[i],betaij=pars$betaij[i],
sigma=sigma,gamma=gamma,rho=rho,k=k,
Ni=Ni,Ei0=Ei0,Ii0=Ii0,Ri0=Ri0,Nj=Nj,Ej0=Ej0,Ij0=Ij0,Rj0=Rj0),
nsim=nsim,format="data.frame",include.data=FALSE
) -> sim_data
#True R_t
sim_data$R_t <- R_t_func(Xi=sim_data$Si, Xj=sim_data$Sj,
betai=pars$betai[i],betaj=pars$betaj[i],betaij=pars$betaij[i],gamma=gamma)
sim_data$.id <- as.numeric(sim_data$.id)
day_below_one = sim_data %>% group_by(.id) %>% filter(R_t<1) %>% filter(row_number()==1)
##EpiEstim R_t
sim_data$reports <- sim_data$reports_i+sim_data$reports_j
sim_data$ince <- sim_data$Ci+sim_data$Cj
T <- length(day)
t_start <- seq(14, T-6) # starting at 2 as conditional on the past observations
t_end <- t_start + 6 # adding 6 to get 7-day windows as bounds included in window
dat_list = split(sim_data, sim_data$.id)
estim_reports_df <- map_dfr(dat_list, ~estimate_R(.x$reports,
method="parametric_si",
config = make_config(list(
t_start = t_start,
t_end = t_end,
mean_si = GI,
std_si = sd_GI)))$R,.id=".id")
estim_reports_df$.id <- as.numeric(estim_reports_df$.id)
day_below_estimR_reports = estim_reports_df %>% group_by(.id) %>%
filter(`Mean(R)`<1)%>% filter(row_number()==1)
sim_data$Day <- as.Date(sim_data$day)
sim_data$week <- as.numeric(strftime(sim_data$Day , format = "%V"))
sim_week <- sim_data%>% group_by(.id,week) %>% summarize(reports=sum(reports)) %>% as.data.frame()
dat_list_week = split(sim_week, sim_week$.id)
agg_week <- map_dfr(dat_list_week,~estimate_R(incid =.x$reports,
dt = 7L,
dt_out = 7L,
recon_opt = "naive",
iter = 10L,
tol = 1e-6,
grid = list(precision = 0.001, min = -1, max = 1),
method="parametric_si",
config = make_config(list(
mean_si = GI,
std_si = sd_GI)))$R,.id=".id")
agg_week <- agg_week[-c(1:6),]
agg_week$.id <- as.numeric(agg_week$.id)
day_below_aggR_report = agg_week %>% group_by(.id) %>%
filter(`Mean(R)`<1)%>% filter(row_number()==1)
thresh_error_reports_agg <- day_below_aggR_report$t_start-day_below_estimR_reports$t_start
sim <- as.factor(rep(pars$run[i],length(thresh_error_reports_agg)))
d_beta <- as.factor(rep(pars$dist[i],length(thresh_error_reports_agg)))
Rt_threshold_list[[i]] <- data.frame(sim,d_beta, thresh_error_reports_agg)
print(i)
}
Rt_threshold_df <- do.call(rbind, Rt_threshold_list)
agg.long <- melt(Rt_threshold_df, id=c("d_beta","sim"))
agg.long$d_beta <- as.factor(agg.long$d_beta)
agg.long$sim <- as.factor(agg.long$sim)
median(agg.long$value)
levels(agg.long$d_beta) <- c("1"=TeX("$\\delta_{\\beta_i} = 2 \\times 10^{-5} $"),
"2"=TeX("$\\delta_{\\beta_i} = 4 \\times 10^{-5} $"),
"3"=TeX("$\\delta_{\\beta_i} = 6 \\times 10^{-5} $"),
"4"=TeX("$\\delta_{\\beta_i} = 8 \\times 10^{-5} $"))
ggplot(agg.long, x=variable, aes(y=value, fill=sim)) +
guides(fill=guide_legend(title=TeX("$\\beta_{ij}$")))+geom_boxplot(outlier.shape = NA)+
scale_y_continuous(limits = c(-5, 10), breaks = seq(-5,10,5))+
scale_fill_brewer(palette="Reds",labels=c('Single Pop.', TeX("$1 \\times 10^{-5}$"),
TeX("$1 \\times 10^{-6}$"),TeX("$1 \\times 10^{-7}$"),
TeX("$1 \\times 10^{-8}$"),"Separate Pops."))+
theme(axis.text.x=element_blank(),
axis.ticks.x = element_blank(),
axis.title.x =element_text(size=14),
axis.title.y =element_text(size=14),
axis.text.y=element_text(size=12),
panel.grid.major.x =element_blank(),
strip.text = element_text(size = 14),
legend.title=element_text(size=12),
legend.text=element_text(size=12))+
ylab(TeX("\\delta$_t$ when \\textit{\\hat{R}}$_t$<1"))+
xlab(TeX("Population Structure $\\rightarrow$"))+theme(legend.text.align = 0)+
facet_wrap(vars(d_beta),labeller = label_parsed, nrow = 1)+
geom_hline(yintercept=0, linetype=3)