Description Usage Arguments Details Value Author(s) Examples
EM algorithm to get MLEs for latent CTMC parameters
1 2 3 4 |
rates.setup |
list with rate setup information |
init.setup |
list with initial distribution setup information |
emission.setup |
list with emission distribution setup information |
the.data |
list with the observed data, with one entry per individual |
num.subjects |
number of individuals in the study |
num.states |
number of latent states in the CTMC |
num.obs.states |
number of observed states |
tol |
convergence tolerance for the EM |
absorb.state |
vector of one or more absorbing latent states |
maxiter=500 |
maximum number of iterations of the EM |
DDO.setup |
setup object for disease driven observation model |
do.DDO |
Indicator (T/F) if there are disease driven observation times in the model? |
time.dep.emission |
(Not supported) Indicator if emission distribution has time dependent covariates. |
Run the expectation-maximization algorithm to get the maximum likelihood estimates for latent CTMC model parameters for multistate disease models. Data are discretely observed disease processes from multiple independent individuals. Times may be non-informative scheduled times or disease-dependent (informative) times. Uses the SQUAREM package for EM acceleration.
params |
MLEs for model parameters: rates/emission/initial/DDO, parameters listed in order of model specification |
details |
Details of the EM
optimization from the |
.
updates |
The iterative oupdates for the parmaeter esitmates at each EM step |
LL.updates |
Updates of observed data likelihood at each step |
time |
Run time |
num.evals |
Number of iterations of the EM |
LL |
Final complete data log likelihood |
Jane Lange
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 | ## Not run:
library(cthmm)
#load the model setup and example data for a competing risks model
#observed at scheduled and informative observation times.
data(DDO_data)
#run the EM on the example data
DDO_EM=EM(rates.setup=rates.setup,
init.setup=init.setup,
emission.setup=emission.setup,
the.data=DDO_data,
num.subjects=500,
num.states=4,
num.obs.states=3,
tol = 1e-07,
absorb.state=c(3,4),
maxiter = 500,
DDO.setup = DDO.setup,
do.DDO = T)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.