Description Usage Arguments Value Author(s) References
This function fits a Gamma-Gompertz model to capture-recapture data using data cloning. To improve parameter estimates clones of the data are used. To speed up convergence at each new number of clones the means and precisions of the priors supposed normally distributed (the last parameter in alphabetical order is logged) are updated. Only for the first number of clones the means must be put in vector initmeans, and a common precision is set in initprec. The specifics of the Markov chains along with the different series of clones are defined
1 2 |
mydata |
a matrix of 1's and 2's for detections and non-detections (the first 1 being the birth year) |
clo |
vector of numbers of clones in each lot: increasing order is advisable and start with 1 clone is not compulsory. Default is c(1,10,50,100) |
nu |
number of updates in the Markov chains. Default is 1000. |
ni |
number of iterations in the Markov chains. Default is 5000. |
nt |
number of thinning (see Link and Eaton 2012). Default is 5. |
nc |
number of Markov chains. Default is 3. |
initmeans |
for the first lot of clones, vector of the normally distributed means of the parameter priors, the last parameter being logged. Default is c(0.15,0.5,4.9,1.4,-0.4). |
initprec |
a common precision of the prior normal distributions is set for all the parameters. Default is 1000. |
This function returns values from the posterior distribution of the parameters of a Gamma-Gompertz model.
a is the coefficient in the baseline mortality rate
b is the coefficient in the exponent of the mortality rate
k is the inverse of the Gamma variance
psi is the mean of the detection rate on the logit scale
sigeta is the standard deviation of the detection rate on the logit scale
Gilbert Marzolin, Olivier Gimenez
Link, W. A. and Eaton, M. J. (2012), On thinning of chains in MCMC. Methods in Ecology and Evolution, 3: 112–115. doi:10.1111/j.2041-210X.2011.00131.x
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.