Abstract
A new procedure for computing the familiar Wilks’ lambda test statistic for testing hypotheses about parameters in the full-rank multivariate regression model is described. Our procedure has the advantage of requiring only one set of computations and possibly one computer run of a standard canonical correlation analysis software package. The new procedure is illustrated with a small data set and the BMDP6M program.
Original language | English (US) |
---|---|
Pages (from-to) | 333-340 |
Number of pages | 8 |
Journal | Journal of Statistical Computation and Simulation |
Volume | 20 |
Issue number | 4 |
DOIs | |
State | Published - Jan 1 1985 |
Externally published | Yes |
ASJC Scopus subject areas
- Statistics and Probability
- Modeling and Simulation
- Statistics, Probability and Uncertainty
- Applied Mathematics