[R] svd (via MFA): All coordinates fall on straight lines?

[Pundurs, Mark]

Applying the svd function to my data by way of the FactoMineR package's MFA function:

dfmfa <- MFA(df, group=c(2,96), type=c("n","c"))

the result is that all my data points fall on one of 8 straight parallel lines when projected onto any two axes, e.g.,

points(dfmfa$ind$coord[, c(1, 2)])

Furthermore, spot-checks indicate that for any pair of axes other than (1, 2) the lines are either horizontal or vertical.

Is all this telling me something about my data, and if so, what? Or is it telling me I should be setting some parameters to nondefault values, and if so, which parameters and what values?

Thanks,
Mark Pundurs
Data Analyst - Traffic
Nokia Location & Commerce, Chicago


The information contained in this communication may be C...{{dropped:8}}