[R-pkgs] Introducing empirical: Probability Distributions as Models of Data

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[R-pkgs] Introducing empirical: Probability Distributions as Models of Data

Abs Spurdle
hi all

I would like to introduce my R package:
empirical: Probability Distributions as Models of Data

The description is:
Computes continuous (not step) empirical (and nonparametric) probability
density, cumulative distribution and quantile functions. Supports
univariate, multivariate and conditional probability distributions, some
kernel smoothing features and weighted data (possibly useful mixed with
fuzzy clustering). Can compute multivariate and conditional probabilities.
Also, can compute conditional medians, quantiles and modes.

Notes:
(1) I'm planning to support categorical variables in the future.
(2) There are some problems with univariate models (but not multivariate),
especially PDFs.
(3) Contrary to what the name empirical suggests, currently multivariate
models use kernel smoothing.
(4) I'm interested in implementing a hybrid Kernel-Quantile method, which I
suspect may be more robust to outliers. If I succeed, then I may rewrite
the univariate implementation.

The URL is:
https://cran.r-project.org/package=empirical

I've written a vignette which describes the package in more detail.
It's URL is:
https://cran.r-project.org/web/packages/empirical/vignettes/empirical.pdf


kind regards
Abs

        [[alternative HTML version deleted]]

_______________________________________________
R-packages mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-packages

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.