On Mar 7, 2013, at 4:47 AM, Kjetil Kjernsmo wrote:
> On Wednesday 6. March 2013 16.33.34 Peter Claussen wrote:
>> But you don't have enough data points to estimate all of the possible
>> interactions; that's why you have NA in your original results.
>
> Yes, but it seems to me that lm is doing
> So, there are at least two points of confusion here, one is
> how coef() differs from effects() in the case of fractional
> factorial experiments, and the other is the factor 1/4
> between the coefficients used by Wu & Hamada and the values
> returned by effects() as I would think from theory
On Wednesday 6. March 2013 14.50.23 Ben Bolker wrote:
>Just a quick thought (sorry for removing context): what happens if
> you use sum-to-zero contrasts throughout, i.e.
> options(contrasts=c("contr.sum", "contr.poly")) ... ?
Ah, I've got it now, this pointed me in the right direction. Thanks
On Thu, Mar 7, 2013 at 5:47 AM, Kjetil Kjernsmo wrote:
> On Wednesday 6. March 2013 16.33.34 Peter Claussen wrote:
>> But you don't have enough data points to estimate all of the possible
>> interactions; that's why you have NA in your original results.
>
> Yes, but it seems to me that lm is doing
On Wednesday 6. March 2013 16.33.34 Peter Claussen wrote:
> But you don't have enough data points to estimate all of the possible
> interactions; that's why you have NA in your original results.
Yes, but it seems to me that lm is doing the right thing, or at least the
expected thing, here, the NA
On Mar 6, 2013, at 4:46 AM, Kjetil Kjernsmo wrote:
> All,
>
> I have just returned to R after a decade of absence, and it is good to see
> that R has become such a great success! I'm trying to bring Design of
> Experiments into some aspects of software performance evaluation, and to
> teach
On Mar 6, 2013, at 9:23 AM, Kjetil Kjernsmo wrote:
> On 03/06/2013 04:18 PM, Peter Claussen wrote:
>> I'll ignore the rest of your question, in the hope that this will answer
>> them sufficiently.
>
> OK!
>
>> You probably want a simple linear model, specified in R using "+" instead of
>> "*
On 03/06/2013 04:18 PM, Peter Claussen wrote:
I'll ignore the rest of your question, in the hope that this will answer them
sufficiently.
OK!
You probably want a simple linear model, specified in R using "+" instead of
"*".
>leaf.lm <- lm(yavg ~ B + C + D + E + Q, data=leaf)
>leaf.lm
Cal
On Mar 6, 2013, at 4:46 AM, Kjetil Kjernsmo wrote:
> All,
>
> I have just returned to R after a decade of absence, and it is good to see
> that R has become such a great success! I'm trying to bring Design of
> Experiments into some aspects of software performance evaluation, and to
> teach
As Ista indicates, the basic issue is that the OP does not understand
linear modeling and is therefore just thrashing around with lm. For
example, the statement about effects being double coefficient is only
true with the orthogonal (-1,1) parameterization of the contrasts.
So I suggest the OP eit
On 03/06/2013 02:50 PM, Ben Bolker wrote:
Just a quick thought (sorry for removing context): what happens if
you use sum-to-zero contrasts throughout, i.e. options(contrasts=c("contr.sum",
"contr.poly")) ... ?
That works (except for the sign)! What would this mean?
Kjetil
Hi,
On Wed, Mar 6, 2013 at 5:46 AM, Kjetil Kjernsmo wrote:
> All,
>
> I have just returned to R after a decade of absence, and it is good to see
> that R has become such a great success! I'm trying to bring Design of
> Experiments into some aspects of software performance evaluation, and to
> tea
Kjetil Kjernsmo ifi.uio.no> writes:
>
> All,
>
> I have just returned to R after a decade of absence, and it is good to
> see that R has become such a great success! I'm trying to bring Design
> of Experiments into some aspects of software performance evaluation, and
> to teach myself that,
All,
I have just returned to R after a decade of absence, and it is good to
see that R has become such a great success! I'm trying to bring Design
of Experiments into some aspects of software performance evaluation, and
to teach myself that, I picked up "Experiments: Planning, Analysis and
Op
14 matches
Mail list logo
