| edr.n {censorSIR} | R Documentation |
take a csir(result of cen.sir) or ds(result of double.slice) object as input, return the effective dimension reduction directions.
edr.n(object, n) edr.percent(object, percent)
object |
a variable of class csir(result of cen.sir) or ds(result of double.slice) |
n |
the number of directions to return |
percent |
the minimum percentage of information that the returned directions should explain, real number between 0 and 1 |
edr.n simply return the first n eigen-vectors of an csir or ds
object. edr.percent simply return the first n eigen-vectors that
explain at least percent information.
To decide how many directions should be used, users can read the
print-out of a csir or a ds object, where result of Chi-Square test and
size of eigen values are shown.
a matrix, each column is one eigen-vector to be used.
These two functions: edr.n and edr.percent do not decide
how many directions should be counted as e.d.r. directions. There information
depend on users input.
Wei Sun sunwei@stat.ucla.edu
Ker-chau Li, Jane-ling Wang and Chun-Hou Chen (1999) Dimension reduction for censored regression data. The Annals of Statistics 27, 1-23.