summary.insilico {InSilicoVA}

Summarizing InSilicoVA Model Fits

Description

This function is the summary method for class insilico.

Usage

## S3 method for class 'insilico'
summary(
  object,
  CI.csmf = 0.95,
  CI.cond = 0.95,
  file = NULL,
  top = 10,
  id = NULL,
  ...
)

Arguments

object

Fitted "insilico" object.

CI.csmf

Confidence interval for CSMF estimates.

CI.cond

Confidence interval for conditional probability estimates

file

Optional .csv file to write to. If it is specified, individual cause of death distribution will be saved to the file.

top

Number of top causes to display on screen.

id

ID of specific death to display on screen.

...

Not used.

Details

summary.insilico formats some basic information about the InSilicoVA fitted object on screen and show the several top CSMFs of user's choice. See below for more detail.

Value

id.all

all IDs of the deaths.

indiv

individual Cause of Death distribution matrix.

csmf

CSMF distribution and confidence interval for each cause.

csmf.ordered

CSMF distribution and confidence interval for each cause, ordered by mean.

condprob

Conditional probability matrix and confidence intervals.

updateCondProb

Component of "insilico" object.

keepProbbase.level

Component of "insilico" object.

datacheck

Component of "insilico" object.

Nsim

Component of "insilico" object.

thin

Component of "insilico" object.

burnin

Component of "insilico" object.

jump.scale

Component of "insilico" object.

levels.prior

Component of "insilico" object.

levels.strength

Component of "insilico" object.

trunc.min

Component of "insilico" object.

trunc.max

Component of "insilico" object.

subpop_counts

Component of "insilico" object.

showTop

Component of "insilico" object.

Author(s)

Zehang Li, Tyler McCormick, Sam Clark

Maintainer: Zehang Li <[e-mail]>

References

Tyler H. McCormick, Zehang R. Li, Clara Calvert, Amelia C. Crampin, Kathleen Kahn and Samuel J. Clark Probabilistic cause-of-death assignment using verbal autopsies, Journal of the American Statistical Association (2016), 111(515):1036-1049.

See Also

insilico, plot.insilico

Examples

## Not run: 
# load sample data together with sub-population list
data(RandomVA1)
# extract InterVA style input data
data <- RandomVA1$data
# extract sub-population information. 
# The groups are "HIV Positive", "HIV Negative" and "HIV status unknown".
subpop <- RandomVA1$subpop

# run without subpopulation
fit1<- insilico( data, subpop = NULL, 
              Nsim = 400, burnin = 200, thin = 10 , seed = 1,
              external.sep = TRUE, keepProbbase.level = TRUE)
summary(fit1)
summary(fit1, top = 10)

# save individual COD distributions to files
summary(fit1, file = "results.csv")

## End(Not run)