[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: how drawing confidence ellipses in a pca?
Luigi Curini --
-findit ellipse- points to -ssc install ellip- and the related article
Alexandersson, A. 1998. gr32: Confidence ellipses. Stata Technical Bulletin
46: 10-13. In Stata Technical Bulletin Reprints, vol. 8, 54-57. College
Station, TX: Stata Press.
thouhg you may have to -ssc install eret2- to save your estimates as
e(b) and e(V) before running -ellip- on "coefficients."
On 3/6/07, Luigi Curini <email@example.com> wrote:
I'd like to have your advice about a puzzle I have. I need to use a
principal component analysis (PCA) to estimate the positions of parties
within a 2-dimensional space. However, to complete the analysis, I'd like to
visualize into the graph of this 2-dimensional (political) world not only
the political parties' means, but also the uncertainty in their mean
positions, drawing confidence ellipses around them (or at least, drawing
confidence segments centered on each mean point along the two dimensions).
This is because parties have widely different standard deviations along the
two dimensions. How to do it?
Thanx for any advice,
* For searches and help try: