strandCet (version 1.0)

HP.priors: Heligman-Pollard Parameter prior formation.

Description

Draws from a uniform distribution with bounds "pri.lo" and "pri.hi" to create the prior distribution of the Heligman-Pollard parameters necessary for the Bayesian Melding procedure.

Usage

HP.priors(pri.lo = c(0, 0, 0, 0.001, 0, 0, 15, 0, 0), pri.hi = c(0.15, 1, 1,
  0.5, 0.25, 15, 55, 0.1, 1.25), theta.dim = 9)

Arguments

pri.lo

Lower bound of the uniform from which the prior is drawn.

pri.hi

Upper bound of the uniform from which the prior is drawn.

theta.dim

The number of parameters to be estimated.

Value

A ((1000 * theta.dim) x theta.dim) matrix containing the 1000 * theta.dim sets of the Heligman-Pollard parameters drawn from a uniform distribution.

References

Heligman, L. and Pollard, J.H. (1980). The Age Pattern of Mortality. Journal of the Institute of Actuaries 107:49<U+2013>80.

Poole, D. and Raftery, A. (2000). Inference for Deterministic Simulation Models: The Bayesian Melding Approach. Journal of the American Statistical Association 95:1244<U+2013>1255.

Raftery, A. and Bao, L. (2009). Estimating and Projecting Trends in HIV/AIDS Generalized Epidemics Using Incremental Mixture Importance Sampling. Technical Report 560, Department of Statistics, University of Washington.

Sharrow, D.J., Clark, S.J., Collinson, M.A., Kahn, K. and Tollman, S.M. (2013). The Age Pattern of Increases in Mortality Affected by HIV: Bayesian Fit of the Heligman-Pollard Model to Data from the Agincourt HDSS Field Site in Rural Northeast South Africa. Demogr. Res. 29, 1039<U+2013>1096.

Examples

# NOT RUN {
priors <- data.frame(priors.lo = c(0,0.5,0,0,0,0,6,0,1),
                     priors.hi = c(0.1,1,1,0.15,0.15,50,10,0.01,1.5))

HP.priors(pri.lo = priors$priors.lo,
          pri.hi = priors$priors.hi,
          theta.dim = 9)

# }