Learn R Programming

SpatioTemporal (version 1.1.7)

est.mesa.model: Examples of estimateSTmodel structure

Description

Example of a model structure holding parameter estimates for the model in mesa.model using estimate.STmodel. Estimation results are also provided for models including spatio-temporal covariates.

Arguments

Format

A list with elements, see the return description in estimate.STmodel.

References

M. A. Cohen, S. D. Adar, R. W. Allen, E. Avol, C. L. Curl, T. Gould, D. Hardie, A. Ho, P. Kinney, T. V. Larson, P. D. Sampson, L. Sheppard, K. D. Stukovsky, S. S. Swan, L. S. Liu, J. D. Kaufman. (2009) Approach to Estimating Participant Pollutant Exposures in the Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air). Environmental Science & Technology: 43(13), 4687-4693.

See Also

estimate.STmodel for parameter estimation. createSTmodel for creation of the originating STmodel object.

Other example data: est.cv.mesa, MCMC.mesa.model, mesa.data.raw, mesa.model, pred.mesa.model

Examples

Run this code
# NOT RUN {
##load a model object
data(mesa.model)

##create vector of initial values
dim <- loglikeSTdim(mesa.model)
x.init <- cbind(c( rep(2, dim$nparam.cov-1), 0),
                c( rep(c(1,-3), dim$m+1), -3, 0))
rownames(x.init) <- loglikeSTnames(mesa.model, all=FALSE)

# }
# NOT RUN {
  ##estimate parameters
  est.mesa.model <- estimate(mesa.model, x.init, hessian.all=TRUE)
# }
# NOT RUN {
##time consuming estimation, load pre-computed results instead
data(est.mesa.model)

#estimation results
print(est.mesa.model)

##compare the estimated parameters for the two starting points
est.mesa.model$summary$par.all
##and values of the likelihood (and convergence info)
est.mesa.model$summary$status

##extract the estimated parameters and approximate uncertainties
x <- coef(est.mesa.model)

##compare estimated parameters
##plot the estimated parameters with uncertainties
par(mfrow=c(1,1),mar=c(13.5,2.5,.5,.5))
with(x, plot(par, ylim=range(c(par-1.96*sd, par+1.96*sd)),
                xlab="", xaxt="n"))
with(x, points(par - 1.96*sd, pch=3))
with(x, points(par + 1.96*sd, pch=3))

abline(h=0, col="grey")
##add axis labels
axis(1, 1:length(x$par), rownames(x), las=2)

# }
# NOT RUN {
  ##example using a few fixed parameters
  x.fixed <- coef(est.mesa.model)$par
  x.fixed[c(1,2,5:9)] <- NA
  est.fix <- estimate(mesa.model, x.init, x.fixed, type="p")
# }

Run the code above in your browser using DataLab