final.resamp: Final re-sampling step in Bayesian Melding using IMIS In HPbayes: Heligman Pollard mortality model parameter estimation using Bayesian Melding with Incremental Mixture Importance Sampling

Description

Performs the final re-sampling step in the Bayesian Melding with IMIS procedure for the eight Heligman-Pollard parameters.

Usage

 ```1 2 3``` ```final.resamp(K, B1, H.new, H.k, log.like, d.keep, prior, h.mu, h.sig, nrisk, ndeath, B = 400, theta.dim = 8, age = c(1e-05, 1, seq(5, 100, 5))) ```

Arguments

 `K` The number of iterations of the importance sampling stage `B1` sample size at the importance sampling stage multiplied by the number of local optimums `H.new` A matrix with dimensions (B*d.keep) x 8 containing the B*d.keep inputs drawn from the multivariate gaussians `H.k` A matrix containing the prior plus new inputs from the multivariate gaussians `log.like` A vector of log-likelihoods corresponding to each row of H.k `d.keep` The number of local optimums found in the optimizer step `prior` A matrix containing the prior `h.mu` A d.keep x 8 matrix containing the results of the optimizer step `h.sig` An array containing the covariance matrix for each row of h.mu `nrisk` A vector containing the number of persons at risk in each age group `ndeath` A vector containing the number of deaths in each age group `B` sample size at the importance sampling stage `theta.dim` The number of columns of the prior matrix `age` A vector containing the ages at which each age interval begins

Details

The function, hp.bm.imis, will perform this along with all other steps in a single function

Value

 `H.new` A `B` x `theta.dim` matrix containing the posterior distribution for each parameter `vwts` A vector containing the variance of the rescaled weights at each IMIS iteration `ewts` A vector containing the entropy of the rescaled weights at each IMIS iteration `mwts` A vector containing the maximum of the rescaled weights at each IMIS iteration `nup` A vector containing the expected number of unique points at each IMIS iteration `frac.up` A vector containing the proportion of unique points in the final resample at each IMIS iteration `wts.k` A vector containing the importance weights for the final iteration `mwt.case` The maximum weight value and associated case

References

Raftery, Adrian and Le Bao. 2009. "Estimating and Projecting Trends in HIV/AIDS Gen- eralized Epidemics Using Incremental Mixture Importance Sampling." Technical Report 560, Department of Statistics, University of Washington.

Poole, David and Adrian Raftery. 2000. "Inference for Deterministic Simulation Models: The Bayesian Melding Approach." Journal of the American Statistical Association 95:1244–1255.

Heligman, Larry and John H. Pollard. 1980 "The Age Pattern of Mortality." Journal of the Institute of Actuaries 107:49–80.

`loop.optim`, `samp.postopt`, `like.resamp`, `hp.bm.imis`, `entropy.wts`, `expt.upts`, `var.rwts`
 ``` 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``` ```## Not run: data(HPprior) lx dx summary(q0) age opt.result <- loop.optim(prior = q0, nrisk=lx, ndeath=dx) opt.mu.d <- opt.result\$opt.mu.d opt.cov.d <- opt.result\$opt.cov.d theta.new <- opt.result\$theta.new d.keep <- opt.result\$d.keep log.like.0 <- opt.result\$log.like.0 wts.0 <- opt.result\$wts.0 samp.po <- samp.postopt(opt.cov.d = opt.cov.d, opt.mu.d = opt.mu.d, prior = q0, d.keep = d.keep) H.k <- samp.po\$H.k H.new <- samp.po\$H.new B1 <- samp.po\$B1 ll.postopt <- like.resamp(K = 10, log.like.0 = log.like.0, opt.cov.d = opt.cov.d, opt.mu.d = opt.mu.d, d.keep = d.keep) h.mu <- ll.postopt\$h.mu h.sig <- ll.postopt\$h.sig log.like <- ll.postopt\$log.like K <- ll.postopt\$K result <- final.resamp(K = K, B1 = B1, H.new = H.new, H.k = H.k, log.like = log.like, d.keep = d.keep, prior = q0, h.mu = h.mu, h.sig = h.sig, nrisk=lx, ndeath=dx, age=age) ## End(Not run) ```