machines: Computer Hardware

In rrcov: Scalable Robust Estimators with High Breakdown Point

Computer Hardware

Description

A data set containing relative CPU performance data of 209 machines on 8 variables. are predictive, one (PRP) is the goal field and one (ERP) is the linear regression's guess. The estimated relative performance values were estimated by the authors using a linear regression method. See their article (Ein-Dor and Feldmesser, CACM 4/87, pp 308-317) for more details on how the relative performance values were set.

Usage

data(machines)

Format

A data frame with 209 rows and 8 variables The variables are as follows:

• MMIN: minimum main memory in kilobytes (integer)

• MMAX: maximum main memory in kilobytes (integer)

• CACH: cache memory in kilobytes (integer)

• CHMIN: minimum channels in units (integer)

• CHMAX: maximum channels in units (integer)

• PRP: published relative performance (integer)

• ERP: estimated relative performance from the original article (integer)

Source

UCI Archive

References

Phillip Ein-Dor and Jacob Feldmesser (1987), Attributes of the performance of central processing units: A relative performance prediction model, Communications of the ACM, 30, 4, pp 308-317.

M. Hubert, P. J. Rousseeuw and T. Verdonck (2009), Robust PCA for skewed data and its outlier map, Computational Statistics & Data Analysis, 53, 2264–2274.

Examples

 data(machines)

 ## Compute the medcouple of each variable of the Computer hardware data
     data.frame(MC=round(apply(machines, 2, mc),2))

 ## Plot a pairwise scaterplot matrix
     pairs(machines[,1:6])

     mcd <- CovMcd(machines[,1:6])
     plot(mcd, which="pairs")

 ##  Remove the rownames (too long)
     rownames(machines) <- NULL

 ## Start with robust PCA based on MCD (P << n)
     (pca1 <- PcaHubert(machines, k=3))
     plot(pca1, main="ROBPCA-MCD", off=0.03)

 ## PCA with the projection algoritm of Hubert
     (pca2 <- PcaHubert(machines, k=3, mcd=FALSE))
     plot(pca2, main="ROBPCA-SD", off=0.03)

 ## PCA with the adjusted for skewness algorithm of Hubert et al (2009)
     (pca3 <- PcaHubert(machines, k=3, mcd=FALSE, skew=TRUE))
     plot(pca3, main="ROBPCA-AO", off=0.03)

rrcov documentation built on July 9, 2023, 6:03 p.m.