openVA (version 1.1.2)

interVA_train: Extended InterVA method for non-standard input

Description

Extended InterVA method for non-standard input

Usage

interVA_train(
  data,
  train,
  causes.train,
  causes.table = NULL,
  thre = 0.95,
  type = c("quantile", "fixed", "empirical")[1],
  prior = c("uniform", "train")[1],
  ...
)

Value

fitted interVA object

Arguments

data

A matrix input, or data read from csv files. Sample input is included as data(RandomVA3).

train

A matrix input, or data read from csv files in the same format as data, but with an additional column specifying cause-of-death. Sample input is included as data(RandomVA3).

causes.train

the column name of the cause-of-death assignment label in training data.

causes.table

list of causes to consider in the training data. Default to be NULL, which uses all the causes present in the training data.

thre

numerical number between 0 and 1. Symptoms with missing rate higher than thre in the training data will be dropped from both training and testing data.

type

type of data conversion when calculating conditional probability (probability of each symptom given each cause of death) for InterVA and InSilicoVA models. Both ``quantile'' and ``fixed'' usually give similar results empirically.

• quantile: the rankings of the P(S|C) are obtained by matching the same quantile distributions in the default InterVA P(S|C)

• fixed: P(S|C) are matched to the closest values in the default InterVA P(S|C) table.

• empirical: no ranking is calculated, but use the empirical conditional probabilities directly.

prior

The prior distribution of CSMF. ``uniform'' uses no prior information, i.e., 1/C for all C causes and ``train'' uses the CSMF in the training data as prior distribution of CSMF.

...

not used

References

Tyler H. McCormick, Zehang R. Li, Clara Calvert, Amelia C. Crampin, Kathleen Kahn and Samuel J. Clark (2016) Probabilistic cause-of-death assignment using verbal autopsies. https://arxiv.org/abs/1411.3042, To appear, Journal of the American Statistical Association

Zehang R. Li, Tyler H. McCormick, Samuel J. Clark (2014) InterVA4: An R package to analyze verbal autopsy data., Center for Statistics and the Social Sciences Working Paper, No.146

http://www.interva.net/

Examples

# \donttest{
data(RandomVA3)
test <- RandomVA3[1:200, ]
train <- RandomVA3[201:400, ]
out <- interVA_train(data = test, train = train, causes.train = "cause", 
                     prior = "train", type = "quantile")
# }