[Rd] Minor quibble with eigen and La.eigen (PR#3221)

[Prof Brian Ripley]

We certainly can.  (svd seems to do this already.)

Could you (next time) supply a test example for the regression suite, 
though?

On Tue, 10 Jun 2003 j.c.rougier at durham.ac.uk wrote:

> Hi everyone,
> 
> It's a very minor point, but could we ensure that eigen and La.eigen
> return a *matrix* for the "vectors" component of the list by including a
> "drop = FALSE", as specified in the help file, ie put
> 
>     list(values = z$values[ord], vectors = if (!only.values) z$vectors[, 
>         ord, drop = FALSE])

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595