Fit a non-population dynamic model using mcmc

dynmodel.mcmc(
  system,
  model,
  evTable,
  inits,
  data,
  fixPars = NULL,
  nsim = 500,
  squared = TRUE,
  seed = NULL
)

Arguments

system: an RxODE object
model: a list of statistical measurement models
evTable: an Event Table object
inits: initial values of system parameters
data: input data
fixPars: fixed system parameters
nsim: number of mcmc iterations
squared: if parameters be squared during estimation
seed: random number seed

Value

A dyn.mcmc object detailing the model fit

Author

Wenping Wang

Examples

# \donttest{

ode <- "
   dose=200;
   pi = 3.1415926535897931;

   if (t<=0) {
      fI = 0;
   } else {
      fI = F*dose*sqrt(MIT/(2.0*pi*CVI2*t^3))*exp(-(t-MIT)^2/(2.0*CVI2*MIT*t));
   }

   C2 = centr/V2;
   C3 = peri/V3;
   d/dt(centr) = fI - CL*C2 - Q*C2 + Q*C3;
   d/dt(peri)  =              Q*C2 - Q*C3;
"
sys1 <- RxODE(model = ode)
#>  


## ------------------------------------------------------------------------
dat <- invgaussian
mod <- cp ~ C2 + prop(.1)
inits <- c(MIT = 190, CVI2 = .65, F = .92)
fixPars <- c(CL = .0793, V2 = .64, Q = .292, V3 = 9.63)
ev <- eventTable()
ev$add.sampling(c(0, dat$time))
(fit <- dynmodel.mcmc(sys1, mod, ev, inits, dat, fixPars))
#> 
#>              mean           sd       cv%
#> MIT  3.951691e+04 5.392877e+03 13.647012
#> CVI2 4.693712e-01 7.590158e-02 16.170908
#> F    8.302230e-01 5.687703e-02  6.850814
#> err1 1.227750e-02 9.331666e-03 76.006208
#> 
#> # samples: 500 
# }