robustbase (version 0.5-0-1)

pulpfiber: Pulp Fiber and Paper Data

Description

Measurements of aspects pulp fibers and the paper produced from them. Four properties of each are measured in sixty-two samples.

Usage

data(pulpfiber)

encoding

utf-8

source

Rousseeuw, P.~J., Van Aelst, S., Van Driessen, K., and Agulló, J. (2004) Robust multivariate regression; Technometrics 46, 293--305.

http://allserv.ugent.be/~svaelst/data/pulpfiber.txt

Details

Cited from the reference article: The dataset contains measurements of properties of pulp fibers and the paper made from them. The aim is to investigate relations between pulp fiber properties and the resulting paper properties. The dataset contains $n = 62$ measurements of the following four pulp fiber characteristics: arithmetic fiber length, long fiber fraction, fine fiber fraction, and zero span tensile. The four paper properties that have been measured are breaking length, elastic modulus, stress at failure, and burst strength.

The goal is to predict the $q = 4$ paper properties from the $p = 4$ fiber characteristics.

References

Lee, J. (1992) Relationships Between Properties of Pulp-Fibre and Paper, unpublished doctoral thesis, U. Toronto, Faculty of Forestry.

Examples

data(pulpfiber)
str(pulpfiber)

pairs(pulpfiber, gap=.1)
## 2 blocks of 4 ..
c1 <- cov(pulpfiber)
cR <- covMcd(pulpfiber)
## how different are they: The robust estimate has more clear high correlations:
symnum(cov2cor(c1))
symnum(cov2cor(cR$cov))