-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathABC.r
190 lines (128 loc) · 4.48 KB
/
ABC.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
library(fitcourseR)
my_SEITL <- SEITL_createModelTdC(deterministic=FALSE)
distanceOscillation
# my summary functions
time.data <- my_SEITL$data$time
inc.data <- my_SEITL$data$Inc
# code
my_summaryStatPeak <- function(time,inc){
# highest peak
return(max(inc))
}
my_summaryStatPeakTime <- function(time,inc){
#
return(time[max(inc)])
}
my_summaryStatFinalSize <- function(time,inc){
# final size
return(sum(inc))
}
my_summaryStatPeak0to20 <- function(time,inc){
# max inc between day 0 and 20
return(max(inc[time<20]))
}
my_summaryStatPeakAfter20 <- function(time,inc){
# max inc between day 0 and 20
return(max(inc[time>20]))
}
my_summaryStatInc0to20 <- function(time,inc){
# inc between day 0 and 20
return(inc[time >= 0 & time <= 20])
}
my_summaryStatInc25to45 <- function(time,inc){
# inc between day 25 and 45
return(inc[time >= 25 & time <= 45])
}
# test
ss1.data <- my_summaryStatPeak(time=time.data,inc=inc.data)
ss2.data <- my_summaryStatPeakTime(time=time.data,inc=inc.data)
ss3.data <- my_summaryStatFinalSize(time=time.data,inc=inc.data)
ss4.data <- my_summaryStatPeak0to20(time=time.data,inc=inc.data)
ss5.data <- my_summaryStatPeakAfter20(time=time.data,inc=inc.data)
ss6.data <- my_summaryStatInc0to20(time=time.data,inc=inc.data)
ss7.data <- my_summaryStatInc25to45(time=time.data,inc=inc.data)
ss.data <- list(ss1.data,ss2.data,ss3.data,ss4.data,ss5.data,ss6.data,ss7.data)
print(ss.data)
simu <- my_SEITL$simulate.model(theta=my_SEITL$theta, state.init=my_SEITL$initialise.state(my_SEITL$theta),times=c(0,my_SEITL$data$time))
simu.obs <- my_SEITL$generate.observation(model.traj=simu, theta=my_SEITL$theta)
time.simu <- simu.obs$time[-1] # remove initial time
inc.simu <- simu.obs$observation[-1] # remove initial state
# test
ss1.simu <- my_summaryStatPeak(time=time.simu,inc=inc.simu)
ss2.simu <- my_summaryStatPeakTime(time=time.simu,inc=inc.simu)
ss3.simu <- my_summaryStatFinalSize(time=time.simu,inc=inc.simu)
ss4.simu <- my_summaryStatPeak0to20(time=time.simu,inc=inc.simu)
ss5.simu <- my_summaryStatPeakAfter20(time=time.simu,inc=inc.simu)
ss6.simu <- my_summaryStatInc0to20(time=time.simu,inc=inc.simu)
ss7.simu <- my_summaryStatInc25to45(time=time.simu,inc=inc.simu)
ss.simu <- list(ss1.simu,ss2.simu,ss3.simu,ss4.simu,ss5.simu,ss6.simu,ss7.simu)
print(ss.simu)
# distances
my_distance1 <- function(data,simu) {
# highest peak
return((data-simu)^2/data)
}
my_distance2 <- function(data,simu) {
# peak time
return(abs(data-simu))
}
my_distance3 <- function(data,simu) {
# final size
return(abs(data-simu))
}
my_distance4 <- function(data,simu) {
# peak first wave
return((data-simu)^2/data)
}
my_distance5 <- function(data,simu) {
# peak second wave
return((data-simu)^2/data)
}
my_distance6 <- function(data,simu) {
# oscillation distance first wave
return(distanceOscillation(data,simu))
}
my_distance7 <- function(data,simu) {
# oscillation distance second wave
return(distanceOscillation(data,simu))
}
# test
d1 <- my_distance1(data=ss1.data,simu=ss1.simu)
d2 <- my_distance2(data=ss2.data,simu=ss2.simu)
d3 <- my_distance3(data=ss3.data,simu=ss3.simu)
d4 <- my_distance4(data=ss4.data,simu=ss4.simu)
d5 <- my_distance5(data=ss5.data,simu=ss5.simu)
d6 <- my_distance6(data=ss6.data,simu=ss6.simu)
d7 <- my_distance7(data=ss7.data,simu=ss7.simu)
print(c(d1,d2,d3,d4,d5,d6,d7))
# acceptance ABC
ABCacceptance <- function(theta,fitmodel,tol) {
# simulate model
simu <- fitmodel$simulate.model(theta=theta, state.init=fitmodel$initialise.state(theta),times=c(0,fitmodel$data$time))
# generate observation
simu.obs <- fitmodel$generate.observation(model.traj=simu, theta=theta)
time.simu <- simu.obs$time[-1] # remove initial time
inc.simu <- simu.obs$observation[-1] # remove initial state
# compute data and simulated summary statistics
# data
time.data <- fitmodel$data$time
inc.data <- fitmodel$data$Inc
ss6.data <- my_summaryStatInc0to20(time=time.data,inc=inc.data)
ss7.data <- my_summaryStatInc25to45(time=time.data,inc=inc.data)
# simu
ss6.simu <- my_summaryStatInc0to20(time=time.simu,inc=inc.simu)
ss7.simu <- my_summaryStatInc25to45(time=time.simu,inc=inc.simu)
# compute distances between summary statistics
d6 <- my_distance6(data=ss6.data,simu=ss6.simu)
d7 <- my_distance7(data=ss7.data,simu=ss7.simu)
# check that distances are within the tolerances
# return 0/1
if(d6<tol[6] && d7 < tol[7]){
return(1)
} else {
return(0)
}
}
for(i in 1:10){
print(ABCacceptance(my_SEITL$theta,my_SEITL,tol=c(0,0,0,0,0,10,10)))
}