Description Usage Arguments Details Value See Also Examples
qibm
is used to fit a Bayesian measurement error model for the
comparison of biomarker measurements derived from different quantitative
imaging methods.
1 2 3 4 |
fixed |
fomula of the form |
image |
quoted or unquoted name of a vector of image identifiers. |
operator |
quoted or unquoted name of a vector of operator identifiers. |
data |
data.frame containing the data vectors. |
priors |
list of prior parameter values. See Details section below for list element names and defaults. |
parameters |
vector naming the model parameters to be returned (default: all). |
n.burnin |
number of MCMC samples to discard as a burn-in sequence. |
n.iter |
total number of samples to generate. |
n.thin |
period at which to save samples. |
n.chains |
number of parallel MCMC chains to generate. |
seed |
numeric random number generator seed. |
Prior distribution hyperprameters may be specified through the following list elements of the priors
argument.
Normal mean for mu parameters (default: 0).
Normal variance for mu paraemters (defaul: 1e6).
Inverse-gamma shape for inter-operator variance parameters (default: 1e-3).
Inverse-gamma rate for inter-operator variance parameters (default: 1e-3).
Two-element vector of uniform minimum and maximum for inter-operator standard deviation parameters (default: inverse-gamma distribution).
Inverse-gamma shape for image-by-operator variance parameters (default: 1e-3).
Inverse-gamma rate for image-by-operator variance parameters (default: 1e-3).
Two-element vector of uniform minimum and maximum for image-by-operator standard deviation parameters (default: inverse-gamma distribution).
Inverse-gamma shape for intra-opearator error variance parameters (default: 1e-3).
Inverse-gamma rate for intra-opearator error variance parameters (default: 1e-3).
Two-element vector of uniform minimum and maximum for intra-opearator standard deviation parameters (default: inverse-gamma distribution).
Scalar pairwise correlation in the correlation matrix defining an inverse-Wishart scale matrix for between-image and method variances (default: 0).
Scalar multiple of the correlation matrix defining an inverse-Wishart scale matrix for between-image and method variances (default: 1).
Inverse-Wishart degrees of freedom for between-image and method variances (default: number of methods).
A qibm
object that inherits from mcmc.list
and
contains the MCMC sampled model parameter values.
describe
,
with
,
Bias
,
CIndex
,
Cor
,
GOF
,
ICC
,
LRM
RC
,
RDC
,
wCV
.
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 | ## Not run:
data(hnc)
## Default prior distributions
fit <- qibm(log(Volume) ~ method, image = lesion, operator = operator,
data = hnc, n.burnin = 5000, n.iter = 10000, n.thin = 5,
n.chains = 3)
describe(with(fit, exp(mu)))
describe(with(fit, exp(sqrt(diag(Sigma.img)))))
describe(with(fit, exp(sqrt(sigma.opr^2 + sigma.imgopr^2))))
describe(with(fit, exp(sigma.err)))
describe(Cor(fit))
describe(Bias(fit, log = TRUE))
describe(CIndex(fit))
describe(ICC(fit))
describe(wCV(fit, log = TRUE))
describe(RDC(fit))
describe(RC(fit))
fit.gof <- GOF(fit)
plot(fit.gof)
describe(fit.gof)
## User-specified Uniform(0, 2) priors on standard deviation parameters
fit2 <- qibm(log(Volume) ~ method, image = lesion, operator = operator,
data = hnc, n.burnin = 5000, n.iter = 10000, n.thin = 5,
n.chains = 3, priors = list(sigma.opr.lim = c(0, 2),
sigma.imgopr.lim = c(0, 2),
sigma.err.lim = c(0, 2)))
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.