A program for finding a robust estimate of shape, location and Mahalanobis distances in high dimension.
This software was used and referenced in the following articles
- Rocke and Woodruff. (1996) Idenfication of Outliers in Multivariate Data. Journal of the American Statistical Association 91:1047-1061.
and
- Markey, Boland and Murphy. (1999) Towards Objective Selection of Representative Microscope Images. Biophysical Journal 76:2230-2237.
To build multout type
make
This should make two files: multout and mulcross.
[icaoberg@lanec1 multout]$ ./multout
multout version 3.03.1
Copyright 1992,93,94,95,96 by David L. Woodruff and David M. Rocke
Usage: multout infile outfile [iterations [parmsfile]]
where infile contains: p n
data record one (p data elements)
data record two
. . .
and iterations is an integer controlling the time spent.
For light contamination use 1 iteration; for heavy, try (n)(p) or more.
If it is not specified, (n)(p) is used.
Example: multout suspect.dat suspect.out 10000
This repository contains an example input file named MULCROSS.DAT. To use this document as an example type
./multout MULCROSS.DAT MULCROSS.OUT 100
You should see output similar to
[icaoberg@lanec1 multout]$ ./multout MULCROSS.DAT MULCROSS.OUT 100
multout version 3.03.1
Copyright 1992,93,94,95,96 by David L. Woodruff and David M. Rocke
Begin Partition Cell 1
Begin Partition Cell 2
Begin Partition Cell 3
Begin Partition Cell 4
Analysis report written to MULCROSS.OUT.
Beginning outlier detection.
Done.
Final report written to MULCROSS.OUT
Feel free to contact the original author, David Woodruff at dlwoodruff AT ucdavis DOT edu
To submit bugs about the source code in this repository please visit
https://github.com/icaoberg/multout
For any other inquiries visit those links as well.
- Port
multoutto python - Make test for
multoutfor python that mimics the C version