
Censored and truncated data
Censored data point are those whose measured
properties are
not known precisely, but are known to lie above or below some limiting
sensitivity. Truncated data points are those which are missing from the
sample altogether due to sensitivity limits.
 ASURV
Astronomy survival analysis for right and leftcensored
data
including the KaplanMeier estimator, several twosample tests,
bivariate correlation measures and linear regressions. Developed for
astronomy by E. Feigelson of Penn State University.
 Partial
correlation for censored data
A test for partial correlation between three variables, any
or
all of which are subject to censoring, based on a generalized Kendall's
tau. Developed for astronomy by M. Akritas (Penn State) and J.
Siebert (MPI).
 Mean
&
standard
deviation
Maximum likelihood estimation of a censored univariate
normal
population. Applied Statistics algorithm #138 distributed by
Statlib.
 Linear
regression
Maximum likelihood multivariate linear regression with
censoring
in the independent variable. Applied Statistics algorithm #139
distributed by Statlib.
 Mean
of
truncated
multivariate dataset
Maximum likelihood estimate of mean and covariance of the
truncated multinormal distribution. Applied Statistics algorithm
#249 distributed by Statlib.
 Mixture
models
for
truncated data
Fits mixture of normals to grouped and truncated
data.
Applied Statistics algorithm #254 distributed by Statlib.
 Multivariate
twosample test with censoring
WeiLachin generalized Wilcoxon and logrank tests for
multivariate data. Applied Statistics algorithm #262 distributed by
Statlib.
 Bounded
variable
least squares
Linear least squares problems with upper and lower bounds
on the
variables. By R. L. Parker (Scripps) and P. B. Stark (UC
Berkeley) and distributed by Statlib.
