Skip to content

table of variance analysis

3 messages · Lecca Paola, Roger Bivand

#
Hello everybody !

I want to evaluate the goodness of the fit of a trend surface. I use the
Least square methods and the function surf.ls. The I write

tma3.sur3 <- surf.ls(3, x, y, z)
summary(tma3.sur3)

Now the question: is the Pr(>F) the probability to obtain a value of
F equal or greater than the obtained one ? Is the confidence level 0.05 ?

Thanks a lot for any help.
Paola.

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
#
On Thu, 21 Mar 2002, Lecca Paola wrote:

            
The value of Pr(>F) is the probability that a value of F of the size
observed could have arisen under the null hypothesis of no relationship
between z and the trend surface in the coordinates (x, y), given the
assumptions made in the model. It is the same as in an anova() of a linear
model object. Confidence levels and probability values are related, but
not the same thing, perhaps a statistics textbook would help?

Roger
#
Sorry in my previous message I express myself not well:

I want to know if the limit for that the value of F is "reasonably
probable" is 0.05.

In other term, when can we say the the two distribution have a
significative disagreement ?

I suppose when Pr(>F) < 0.05.
--Paola.





-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._