readingError-class {mspath} | R Documentation |
Describes errors in readings for discrete states. This is
a special type of matrix
in which rows are the true states and
columns are observed states. The diagonal is mostly ignored, since it
must be 1 - sum of other elements on the row.
Objects can be created by calls of the form readingError(...)
where the arguments are the usual ones for matrix
.
.Data
:"matrix"
Class "matrix"
, from data part.
Class "structure"
, by class "matrix", distance 2.
Class "array"
, by class "matrix", distance 2.
Class "vector"
, by class "matrix", distance 3, with explicit coerce.
Class "vector"
, by class "matrix", distance 4, with explicit coerce.
signature(object = "readingError")
: returns a
matrix with 1's on off-diagonal positive elements, 0 elsewhere. signature(object = "readingError")
: return a
matrix with|code{TRUE} on off-diagonal positive elements,
FALSE
elsewhere.signature(object = "readingError")
: returns a
vector with only the off-diagonal positive elements, starting with
the first row, then the second, and so on.signature(x = "readingError")
: The arguments are
one or more readingError
's. This returns a
readingError
whose entries are the means of each of the
cells in the arguments. Note that extra arguments to mean
elsewhere in R provide optional tuning values for the
calculation; that is not the case here.
This class is designed to work with mspath
and may not be
appropriate for other uses. In particular, coef
returns values
in the order expected by the inits
argument to
mspath
, which is the transpose of the usual order for R.
Ross Boylan
Bacchetti, Peter and Boylan, Ross (2009) “Estimating Complex Multi-State Misclassification Rates for Biopsy-Measured Liver Fibrosis in Patients with Hepatitis C,” The International Journal of Biostatistics: Vol. 5 : Iss. 1, Article 5. DOI: 10.2202/1557-4679.1139 http://www.bepress.com/ijb/vol5/iss1/5
matrix
, mspath
, and sample data
provided with this package.
library(mspath) re <- readingError(c(.73, .27, 0, 0, 0, .02, .73, .25, 0, 0, 0, .02, .90, .08, 0, 0, 0, 0, .88, .13, 0, 0, 0, .07, .93), byrow=TRUE, nrow=5, ncol=5) coef(re)