bayesHistogram.help {bayesSurv} | R Documentation |
Helping function for Bayesian smoothing of (bi)-variate densities based on possibly censored data
Description
These functions are not to be called by ordinary users.
These are just sub-parts of ‘bayesHistogram’ function to make it more readable for the programmer.
Usage
bayesHistogram.design(y1, y2)
bayesHistogram.checkStore(store)
bayesHistogram.priorInit(prior, init, mcmc.par, design)
bayesHistogram.writeHeaders(dir, design, prior.init, store)
Arguments
y1 |
response for the first dimension. This should be a~survival object
created by |
y2 |
response object for the second dimension (if bivariate
density is to be smoothed). This should be a~survival object created
by |
store |
a~list with appropriate components |
prior |
a~list as required by |
init |
a~list as required by |
mcmc.par |
a~list as required by |
design |
a~list with the design information as returned by the
function
|
dir |
string giving a~directory where to store sampled files |
prior.init |
an~object as returned by |
Value
Some lists.
Value for bayesHistogram.design
A~list with the following components:
- y.left
vector or matrix with either observed, right or left censored observations or with the lower limits of interval censored observations. It is a vector if
dim == 1
and it is a matrix with 2 rows andn
columns ifdim == 2
, wheren
is a~sample size. In that case, the first row of the matrix are responses for the first dimension and the second row of the matrix are responses for the second dimension.- y.right
vector or matrix with entries equal to 1 for observed, right or left censored observations and entries equal to the upper limits of interval censored observations. The structure is the same as that of
y.left
.- status
a~vector or matrix with censoring indicators (1 = exactly observed, 0 = right censored, 2 = left censored, 3 = interval censored). The structure is the same as that of
y.left
.- dim
dimension of the response, i.e. 1 (univariate smoothing) or 2 (bivariate smoothing)
Value for bayesHistogram.priorInit
A~list with the following components:
- Gparmi
integer parameters for the G-spline constructor in the C++ code
- Gparmd
double parameters for the G-spline constructor in the C++ code
- iter
index of the nullth iteration
- y
vector of initial values for the response, sorted as
y_1[1], y_1[2], \dots, y_n[1], y_n[2]
in the case of bivariate response with sample size equal ton
- r
initial component labels (vector of size
n
) taking values from 1 to the total length of the G-spline- specification
specification of the G-spline model (1 or 2), see
bayesHistogram
for more detail
and the following attributes:
init |
prior |
mcmc.par |
Author(s)
Arnošt Komárek arnost.komarek@mff.cuni.cz