[S] (Fwd) Q: variables in output from lm

Daniel Svozil (D.Svozil@plymouth.ac.uk)
Fri, 3 Jul 1998 09:19:46 BST


Hello,

I have a data set Pt214.fr consisting of 8 columns obtained from
experimental design (partial factorial design 3^(8-4) and one column
of response. The column names are Pt, Pd,..., response name is
Pt214.423. When I run summary(lm(Pt214.423~Pt+Pd, Pt214.fr)), I
receive the following output:
----------------------------------------------------------------
Call: lm(formula = Pt214.423 ~ Pt + Pd, data = Pt214.fr)
Residuals:
Min 1Q Median 3Q Max
-2248 -538.9 -1.478 760.2 2412

Coefficients:
Value Std. Error t value Pr(>|t|)
(Intercept) 10029.2689 113.7829 88.1439 0.0000
Pt1 2752.9144 136.8663 20.1139 0.0000
Pt2 3995.8466 81.8683 48.8082 0.0000
Pd1 163.8093 136.8663 1.1969 0.2347
Pd2 -43.4677 81.8683 -0.5309 0.5969

Residual standard error: 1060 on 84 degrees of freedom
Multiple R-Squared: 0.9696
F-statistic: 669.1 on 4 and 84 degrees of freedom, the p-value is 0
---------------------------------------------------------------------
I feel a little bit confused, what does it mean Pt1, Pt2, Pd1,
Pd2?????? Which variables are they???

Maybe I should mention, that Pt214.fr consist of 81 experiments
received from experimental desin (i.e. these experiments are
different each to other) and 8 middle-point experiments, that are the
same (each factor is on the middle level).

Thanks in advance
Dan

---------------------------------------
Daniel Svozil
Deparment of Environmental Sciences
University of Plymouth
Drake Circus
Plymouth
Devon PL4 8AA
United Kingdom

e-mail: dsvozil@plymouth.ac.uk
Tel: ++44-(0)1752-23 30 29
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news