jointCureModel: Estimation of the joint longitudinal time-to-event models...
In AntoineBbi/JMcuR: Joint Modeling for longitudinal and time-to-event data with cuRe fraction (JMcuR)
Description
Usage
Arguments
Value
Author(s)
References
Examples
View source: R/jointCureModel.R
Description
Function using JAGS to estimate the joint longitudinal survival models with cure fraction:
mixutre cure model (MCmodel) [Farewell, 1982], joint latent class cure model (JLCCmodel),
full joint cure model (FJCmodel) [Law et al. 2002, Yu et al. 2004,2008].
This function is an extansion (or modification) to the cure framework of the jointModelBayes function from the JMbayes package
(version 0.4-1 implemented by Dimitris Rizopoulos).
Usage
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
jointCureModel(
formFixed,
formRandom,
formLatency,
formIncidence,
formID,
IdVar,
data,
timeVar,
jointCureModel = "FJCmodel",
survMod = "weibull-PH",
param = "shared-RE",
Infprior_cure = F,
smcure_out = NULL,
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = FALSE,
n.chains = 1,
n.iter = 10000,
n.burnin = 5000,
n.thin = 1,
n.adapt = 5000,
C = 1,
priorTau = 100,
out_data = T
)
Arguments
formFixed
formula for fixed part of longitudinal submodel with response variable
formRandom
formula for random part of longitudinal submodel without response variable
formLatency
survival formula as formula in survival package for latency submodel
formIncidence
formula specifying covariate in incidence submodel
formID
formula specifying the ID variable (e.g. = ~ subject)
IdVar
string specify the names of ID variable (Id subject)
data
dataset of observed variables
timeVar
string specify the names of time variable (time of repeated measurements)
jointCureModel
joint cure model to estimate including: "MCmodel","JLCCmodel","FJCmodel"
survMod
form of survival submodel (only "weibull-PH" is available until now)
param
shared association for "FJCmodel": "shared-RE" (default) or "td-value"
Infprior_cure
most of prior distributions are data-driven if 'TRUE', otherwise vague priors are considered
smcure_out
estimated model from smcure package, otherwise is NULL by default
classif_trick
if TRUE, subjects are cured if t_obs>max(t_obs(delta==1), otherwise D is only know for subjects having developed the event Taylor's rule about the cure status, D==0 if T_obs>max(t_obs[which(delta==1)])
cov_prior
="wishart" prior distribtion for random effect covariance matrix or ="inverse-gamma" for independent prior distribtions with random effect variance matrix
Sigma_d
if TRUE, the random effect matrix is class-specific, otherwise the matrix is defined for the whole population
n.chains
the number of parallel chains for the model; default is 1.
n.iter
nteger specifying the total number of iterations; default is 10000
n.burnin
integer specifying how many of n.iter to discard as burn-in ; default is 5000
n.thin
integer specifying the thinning of the chains; default is 1
n.adapt
integer specifying the number of iterations to use for adaptation; default is 5000
C
positive integer for the zero trick used to define the likelihood in JAGS, default is 1
priorTau
variance by default for vague prior distribution
out_data
Boolean such as TRUE if you want the data of different submodels in output or FALSE otherwise
Value
A JMcuR object which is a list with the following elements:
Coefficientslist of posterior mean of each parameter
Modes list of posterior modes compute from the posterior samples of the parameters
Sdlist of Standard deviation of each parameters
sims.listlist of the MCMC chains of the parameters
Author(s)
Antoine Barbieri and Catherine Legrand
References
Barbieri A and Legrand C. (2019). Joint longitudinal and time-to-event cure models for the assessment of being cured. Statistical Methods in Medical Research. doi: 10.1177/0962280219853599.
Proust-Lima C, Philipps V and Liquet B. (2017). Estimation of Extended Mixed Models Using Latent Classes and Latent Processes: The R Package lcmm. Journal of Statistical Software; 78(2).
Rizopoulos, D. (2016). The R Package JMbayes for Fitting Joint Models for Longitudinal and Time-to-Event Data Using MCMC. Journal of Statistical Software, 72(7), 1-46.
Yu M, Taylor JMG and Sandler HM. (2008). Individual Prediction in Prostate Cancer Studies Using a Joint Longitudinal Survival-Cure Model. Journal of the American Statistical Association; 103(481): 178-187.
Yu M, Law NJ, Taylor JMG et al. (2004). Joint longitudinal-survival-cure models and their application to prostate cancer. Statistica Sinica; 14(3): 835-862.
Law NJ, Taylor JMG and Sandler H. (2002). The joint modeling of a longitudinal disease progression marker and the failure time process in the presence of cure. Biostatistics; 3(4): 547-563.
Taylor JMG. (1995) Semi-Parametric Estimation in Failure Time Mixture Models. Biometrics 1995; 51(3): 899-907.
Examples
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
## For the exemple(s), use the data 'aids' from joineR package
data("aids", package = "joineR")
## estimation of the MCM for parameter initialisation
aids.id <- unique(aids[,c("id","time","death","drug","gender","prevOI","AZT")])
aids.id2 <- aids.id
aids.id2$drug <- as.numeric(aids.id$drug)-1
aids.id2$gender <- as.numeric(aids.id$gender)-1
aids.id2$prevOI <- as.numeric(aids.id$prevOI)-1
aids.id2$AZT <- as.numeric(aids.id$AZT)-1
smcure_out <- smcure(Surv(time, death) ~ drug + gender + prevOI + AZT,
cureform=~ drug + gender + prevOI + AZT,
data = aids.id2,
model="ph")
## Estimation of the joint latent class cure model (JLCCM)
JLCCM <- jointCureModel(formFixed = CD4 ~ obstime + drug + gender + prevOI + AZT,
formRandom = ~ obstime,
timeVar= "obstime",
formLatency = Surv(time, death) ~ drug + gender + prevOI + AZT,
formIncidence = ~ drug + gender + prevOI + AZT,
formID= ~ id,
IdVar = "id",
data = aids,
# model specifications
jointCureModel = "JLCCmodel",
survMod = "weibull-PH",
param = "shared-RE",
# prior options
n.iter = 1000,
n.burnin = 500,
Infprior_cure = TRUE,
smcure_out = smcure_out,
priorTau = 100,
# classification options
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = TRUE)
## details of the estimated model
summary(JLCCM)
## Not run:
## Estimation of the full joint cure model given shared curent value (FJCM_A1)
FJCM_A1 <- jointCureModel(formFixed = CD4 ~ obstime,
formRandom = ~ obstime,
timeVar= "obstime",
formLatency = Surv(time, death) ~ drug + gender + prevOI + AZT,
formIncidence = ~ drug + gender + prevOI + AZT,
formID= ~ id,
IdVar = "id",
data = aids,
# model specifications
jointCureModel = "FJCmodel",
survMod = "weibull-PH",
param = "td-value",
# prior options
Infprior_cure = TRUE,
smcure_out = smcure_out,
priorTau = 100,
# classification options
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = TRUE,
# MCMC options
n.chains = 1, n.iter = 10000, n.burnin = 5000, n.thin = 1, n.adapt = 5000, C = 1,
# out option
out_data=T)
## details of the estimated model
summary(FJCM_A1)
## Estimation of the full joint cure model given shared curent value (FJCM_A1)
FJCM_A2 <- jointCureModel(formFixed = CD4 ~ obstime,
formRandom = ~ obstime,
timeVar= "obstime",
formLatency = Surv(time, death) ~ drug + gender + prevOI + AZT,
formIncidence = ~ drug + gender + prevOI + AZT,
formID= ~ id,
IdVar = "id",
data = aids,
# model specifications
jointCureModel = "FJCmodel",
survMod = "weibull-PH",
param = "shared-RE",
# prior options
Infprior_cure = TRUE,
smcure_out = smcure_out,
priorTau = 100,
# classification options
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = TRUE,
# MCMC options
n.chains = 1, n.iter = 10000, n.burnin = 5000, n.thin = 1, n.adapt = 5000, C = 1,
# out option
out_data=T)
## details of the estimated model
summary(FJCM_A2)
## End(Not run)
AntoineBbi/JMcuR documentation built on Oct. 1, 2020, 1:31 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description Usage Arguments Value Author(s) References Examples
View source: R/jointCureModel.R
Description
Function using JAGS to estimate the joint longitudinal survival models with cure fraction:
mixutre cure model (MCmodel) [Farewell, 1982], joint latent class cure model (JLCCmodel),
full joint cure model (FJCmodel) [Law et al. 2002, Yu et al. 2004,2008].
This function is an extansion (or modification) to the cure framework of the jointModelBayes function from the JMbayes package
(version 0.4-1 implemented by Dimitris Rizopoulos).
Usage
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 | jointCureModel(
formFixed,
formRandom,
formLatency,
formIncidence,
formID,
IdVar,
data,
timeVar,
jointCureModel = "FJCmodel",
survMod = "weibull-PH",
param = "shared-RE",
Infprior_cure = F,
smcure_out = NULL,
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = FALSE,
n.chains = 1,
n.iter = 10000,
n.burnin = 5000,
n.thin = 1,
n.adapt = 5000,
C = 1,
priorTau = 100,
out_data = T
)
|
Arguments
formFixed |
formula for fixed part of longitudinal submodel with response variable |
formRandom |
formula for random part of longitudinal submodel without response variable |
formLatency |
survival formula as formula in survival package for latency submodel |
formIncidence |
formula specifying covariate in incidence submodel |
formID |
formula specifying the ID variable (e.g. = ~ subject) |
IdVar |
string specify the names of ID variable (Id subject) |
data |
dataset of observed variables |
timeVar |
string specify the names of time variable (time of repeated measurements) |
jointCureModel |
joint cure model to estimate including: "MCmodel","JLCCmodel","FJCmodel" |
survMod |
form of survival submodel (only "weibull-PH" is available until now) |
param |
shared association for "FJCmodel": "shared-RE" (default) or "td-value" |
Infprior_cure |
most of prior distributions are data-driven if 'TRUE', otherwise vague priors are considered |
smcure_out |
estimated model from smcure package, otherwise is NULL by default |
classif_trick |
if TRUE, subjects are cured if t_obs>max(t_obs(delta==1), otherwise D is only know for subjects having developed the event Taylor's rule about the cure status, D==0 if T_obs>max(t_obs[which(delta==1)]) |
cov_prior |
="wishart" prior distribtion for random effect covariance matrix or ="inverse-gamma" for independent prior distribtions with random effect variance matrix |
Sigma_d |
if TRUE, the random effect matrix is class-specific, otherwise the matrix is defined for the whole population |
n.chains |
the number of parallel chains for the model; default is 1. |
n.iter |
nteger specifying the total number of iterations; default is 10000 |
n.burnin |
integer specifying how many of n.iter to discard as burn-in ; default is 5000 |
n.thin |
integer specifying the thinning of the chains; default is 1 |
n.adapt |
integer specifying the number of iterations to use for adaptation; default is 5000 |
C |
positive integer for the zero trick used to define the likelihood in JAGS, default is 1 |
priorTau |
variance by default for vague prior distribution |
out_data |
Boolean such as TRUE if you want the data of different submodels in output or FALSE otherwise |
Value
A JMcuR object which is a list with the following elements:
Coefficientslist of posterior mean of each parameter
Modeslist of posterior modes compute from the posterior samples of the parameters
Sdlist of Standard deviation of each parameters
sims.listlist of the MCMC chains of the parameters
Author(s)
Antoine Barbieri and Catherine Legrand
References
Barbieri A and Legrand C. (2019). Joint longitudinal and time-to-event cure models for the assessment of being cured. Statistical Methods in Medical Research. doi: 10.1177/0962280219853599.
Proust-Lima C, Philipps V and Liquet B. (2017). Estimation of Extended Mixed Models Using Latent Classes and Latent Processes: The R Package lcmm. Journal of Statistical Software; 78(2).
Rizopoulos, D. (2016). The R Package JMbayes for Fitting Joint Models for Longitudinal and Time-to-Event Data Using MCMC. Journal of Statistical Software, 72(7), 1-46.
Yu M, Taylor JMG and Sandler HM. (2008). Individual Prediction in Prostate Cancer Studies Using a Joint Longitudinal Survival-Cure Model. Journal of the American Statistical Association; 103(481): 178-187.
Yu M, Law NJ, Taylor JMG et al. (2004). Joint longitudinal-survival-cure models and their application to prostate cancer. Statistica Sinica; 14(3): 835-862.
Law NJ, Taylor JMG and Sandler H. (2002). The joint modeling of a longitudinal disease progression marker and the failure time process in the presence of cure. Biostatistics; 3(4): 547-563.
Taylor JMG. (1995) Semi-Parametric Estimation in Failure Time Mixture Models. Biometrics 1995; 51(3): 899-907.
Examples
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 | ## For the exemple(s), use the data 'aids' from joineR package
data("aids", package = "joineR")
## estimation of the MCM for parameter initialisation
aids.id <- unique(aids[,c("id","time","death","drug","gender","prevOI","AZT")])
aids.id2 <- aids.id
aids.id2$drug <- as.numeric(aids.id$drug)-1
aids.id2$gender <- as.numeric(aids.id$gender)-1
aids.id2$prevOI <- as.numeric(aids.id$prevOI)-1
aids.id2$AZT <- as.numeric(aids.id$AZT)-1
smcure_out <- smcure(Surv(time, death) ~ drug + gender + prevOI + AZT,
cureform=~ drug + gender + prevOI + AZT,
data = aids.id2,
model="ph")
## Estimation of the joint latent class cure model (JLCCM)
JLCCM <- jointCureModel(formFixed = CD4 ~ obstime + drug + gender + prevOI + AZT,
formRandom = ~ obstime,
timeVar= "obstime",
formLatency = Surv(time, death) ~ drug + gender + prevOI + AZT,
formIncidence = ~ drug + gender + prevOI + AZT,
formID= ~ id,
IdVar = "id",
data = aids,
# model specifications
jointCureModel = "JLCCmodel",
survMod = "weibull-PH",
param = "shared-RE",
# prior options
n.iter = 1000,
n.burnin = 500,
Infprior_cure = TRUE,
smcure_out = smcure_out,
priorTau = 100,
# classification options
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = TRUE)
## details of the estimated model
summary(JLCCM)
## Not run:
## Estimation of the full joint cure model given shared curent value (FJCM_A1)
FJCM_A1 <- jointCureModel(formFixed = CD4 ~ obstime,
formRandom = ~ obstime,
timeVar= "obstime",
formLatency = Surv(time, death) ~ drug + gender + prevOI + AZT,
formIncidence = ~ drug + gender + prevOI + AZT,
formID= ~ id,
IdVar = "id",
data = aids,
# model specifications
jointCureModel = "FJCmodel",
survMod = "weibull-PH",
param = "td-value",
# prior options
Infprior_cure = TRUE,
smcure_out = smcure_out,
priorTau = 100,
# classification options
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = TRUE,
# MCMC options
n.chains = 1, n.iter = 10000, n.burnin = 5000, n.thin = 1, n.adapt = 5000, C = 1,
# out option
out_data=T)
## details of the estimated model
summary(FJCM_A1)
## Estimation of the full joint cure model given shared curent value (FJCM_A1)
FJCM_A2 <- jointCureModel(formFixed = CD4 ~ obstime,
formRandom = ~ obstime,
timeVar= "obstime",
formLatency = Surv(time, death) ~ drug + gender + prevOI + AZT,
formIncidence = ~ drug + gender + prevOI + AZT,
formID= ~ id,
IdVar = "id",
data = aids,
# model specifications
jointCureModel = "FJCmodel",
survMod = "weibull-PH",
param = "shared-RE",
# prior options
Infprior_cure = TRUE,
smcure_out = smcure_out,
priorTau = 100,
# classification options
classif_trick = TRUE,
cov_prior = "inverse-gamma",
Sigma_d = TRUE,
# MCMC options
n.chains = 1, n.iter = 10000, n.burnin = 5000, n.thin = 1, n.adapt = 5000, C = 1,
# out option
out_data=T)
## details of the estimated model
summary(FJCM_A2)
## End(Not run)
|
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.