With a matris response lm(...) doesn't really fit a multivariate
regression. It fits multiple univariate linear regressions. You can
build a multivariate linear regression from it, of course.
I suspect you need to looking at manova(...) to fit the model. Then the
summary method, and the other model comparison tools of course, will
give you multivariate tests.
As to simultaneous prediction intervals and ellipsoids, the only advice
I can give you is the standard: fear God, honour the king and work very
hard. Best of luck! :-)
Bill Venables.
-----Original Message-----
From: s-news-owner@lists.biostat.wustl.edu
[mailto:s-news-owner@lists.biostat.wustl.edu] On Behalf Of Ray Haraf
Sent: Sunday, 30 March 2008 3:32 PM
To: s-news@lists.biostat.wustl.edu
Subject: [S] summary(object, test=c("Roy", "Wilks", "Pillai", ....)
Dear All,
I am running multivariate multiple regression using lm() and would be
very appreciative of your kind and prompt help (as usual) with how to
obtain the usual statistics of multivariate analysis, i.e., Wilks'
lambda, Pillai's trace, Hotelling-Lawley trace, and Roy's greatest root.
I unsuccessfully tried this
> ex7.8 <- data.frame(z1 = c(0, 1, 2, 3, 4),
+ y1 = c(1, 4, 3, 8, 9),
+ y2 = c(-1, -1, 2, 3, 2))
> summary(lm(cbind(y1, y2) ~ z1, data =
ex7.8),test=c("Wilks","Roy","Pillai"))
Also, any suggestion on how obtain simultaneous prediction intervals and
prediction ellipsoid would be very valued. Thanks for your kind care
Here is the result of the above summary() function
Response y1 :
Call:
lm(formula = y1 ~ z1, data = ex7.8)
Residuals:
1 2 3 4 5
6.880e-17 1.000e+00 -2.000e+00 1.000e+00 -1.326e-16
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.0000 1.0954 0.913 0.4286
z1 2.0000 0.4472 4.472 0.0208 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 1.414 on 3 degrees of freedom
Multiple R-squared: 0.8696, Adjusted R-squared: 0.8261
F-statistic: 20 on 1 and 3 DF, p-value: 0.02084
Response y2 :
Call:
lm(formula = y2 ~ z1, data = ex7.8)
Residuals:
1 2 3 4 5
9.931e-17 -1.000e+00 1.000e+00 1.000e+00 -1.000e+00
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -1.0000 0.8944 -1.118 0.3450
z1 1.0000 0.3651 2.739 0.0714 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 1.155 on 3 degrees of freedom
Multiple R-squared: 0.7143, Adjusted R-squared: 0.619
F-statistic: 7.5 on 1 and 3 DF, p-value: 0.07142
>
Kind regards,
Ray
|