Description Usage Arguments Details Value Author(s)
Decay fit with modulation of mean depth by Gaussian Process
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 | fitExpGP(
x,
y,
uy,
dataType = 2,
Nn = 10,
theta0 = NULL,
Sigma0 = NULL,
lambda_rate = 0.1,
lasso = FALSE,
method = "sample",
iter = 50000,
prior_PD = 0,
alpha_scale = 0.1,
rho_scale = 1/Nn,
gridType = "internal",
nb_chains = 4,
nb_warmup = 500,
nb_iter = 1000,
verbose = FALSE,
open_progress = TRUE
)
|
x |
a numeric vector |
y |
a numeric vector of responses |
uy |
a numeric vector of uncertainty on 'y' |
dataType |
an numeric (1 or 2) defining the type of data |
Nn |
number of control points |
theta0 |
theta prior mean vbalues |
Sigma0 |
theta prior corvariance matrix |
lambda_rate |
scale of ctrl points prior |
lasso |
flag to use lasso prior |
method |
choice of optimization method in c('optim','sample') |
iter |
max. number of iterations for 'optim' |
prior_PD |
flag to sample from prior pdf only |
alpha_scale |
SD scale of GP |
rho_scale |
relative correlation length of GP |
gridType |
type of controle points grid ('internal' does not contain boundaries) |
nb_chains |
number of MCMC chains |
nb_warmup |
number of warmup steps |
nb_iter |
number of steps |
Bayesian inference of the parameters of an exponential
model with modulation assuming an uncorrelated normal noise
y(x) ~ normal(m(x),uy(x));
m(x) = theta[1] + theta[2]*exp(-dataType*x/theta[3]*(1+l(x)));
.
l(x)
is defined by a GP with fixed positions 'xGP',
variance and correlation length.
A list containing
a stanfit
object containg the results of the fit
a vector of coordinates for the control points
same as input
same as input
same as input
Pascal PERNOT
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.