G'day Gabor,
On Thu, 17 Mar 2011 20:38:21 -0400
Gabor Grothendieck wrote:
> > Or am I missing something?
O.k., because the residuals don't add to zero, there may be a non-zero
correlation between residuals and fitted values, which messes up the
equation at the variance level.
> Try it on an ex
On Thu, Mar 17, 2011 at 8:22 PM, Berwin A Turlach
wrote:
> G'day Gabor,
>
> On Thu, 17 Mar 2011 11:36:38 -0400
> Gabor Grothendieck wrote:
>
>> The idea is that if you have a positive quantity that can be broken
>> down into two nonnegative quantities: X = X1 + X2 then it makes sense
>> to ask wh
G'day Gabor,
On Thu, 17 Mar 2011 11:36:38 -0400
Gabor Grothendieck wrote:
> The idea is that if you have a positive quantity that can be broken
> down into two nonnegative quantities: X = X1 + X2 then it makes sense
> to ask what proportion X1 is of X. For example: 10 = 6 + 4 and 6 is
> .6 of
Thank you for our reply. It's a pity, that 2 variables defined by different
formula have same name. If the variables had been named differently, I
wouldn't have problem at all and it looks like it's done on purpose.
Because I test a quality of data (performance of collecting data) not a
model which
@r-project.org
> Subject: Re: [R] Strange R squared, possible error
>
> Thats exactly what I would like to do. Any idea on good text? I've
> consulted
> severel texts, but no one defined R^2 as R^2 = 1 - Sum(R[i]^2) /
> Sum((y[i])^2-y*)) still less why to use different formul
Thats exactly what I would like to do. Any idea on good text? I've consulted
severel texts, but no one defined R^2 as R^2 = 1 - Sum(R[i]^2) /
Sum((y[i])^2-y*)) still less why to use different formulas for similar model
or why should be R^2 closer to 1 when y=a*x+0 than in general model y=a*x+b.
fr
Yes they are. I had edited the reply, but It didn't help.
Correction:
2)I meant zero slope, no zero intercept.
--
View this message in context:
http://r.789695.n4.nabble.com/Strange-R-squared-possible-error-tp3382818p3384648.html
Sent from the R help mailing list archive at Nabble.com.
Thank you for your very comprehensible answer.
I a priori know that model y=a*x+0 is right and that I can't get x=constant
nor y=constant.
I'm comparing performance of data gathering in my data set to another data
sets in which performance gathering is characterized by R-squared . The data
in dat
On Wed, Mar 16, 2011 at 3:49 PM, derek wrote:
> k=lm(y~x)
> summary(k)
> returns R^2=0.9994
>
> lm(y~x) is supposed to find coef. a anb b in y=a*x+b
>
> l=lm(y~x+0)
> summary(l)
> returns R^2=0.9998
> lm(y~x+0) is supposed to find coef. a in y=a*x+b while setting b=0
>
> The question is why do I g
On 2011-03-17 02:08, derek wrote:
Exuse me, I don't claim R^2 can't be negative. What I say if I get R^2
negative then the data are useless.
I know, that what Thomas said is true in general case. But in my special
case of data, using nonzero intercept is nonsense, and to get R^2 less than
0.985 i
> 2) I don't want to fit data with linear model of zero intercept.
> 3) I dont know if I understand correctly. Im 100% sure the model for my data
> should have zero intercept.
> The only coordinate which Im 100% sure is correct. If I had measured quality
> Y of a same sample X0 number of times I wo
Exuse me, I don't claim R^2 can't be negative. What I say if I get R^2
negative then the data are useless.
I know, that what Thomas said is true in general case. But in my special
case of data, using nonzero intercept is nonsense, and to get R^2 less than
0.985 is considered poor job (standard R^2>
On Wed, Mar 16, 2011 at 4:00 PM, derek wrote:
> 1) In my very humble opinion R^2 can't be negative, at least for data for
> which it sound to use linear model.
!!! Your opinion, humble or not, counts for nothing. Thomas stated a
mathematical fact. I suggest you make an effort to understand what h
On 11-03-16 07:55 PM, Peter Ehlers wrote:
> On 2011-03-16 15:02, Ben Bolker wrote:
>> ria.buffalo.edu> writes:
>>
>>>
>>> lm(y~x+0) yields the regression on x without the constant, i.e., y=bx+e,
>>> not y = a +e
>>>
>>> derek gmail.com>
>>> Sent by: r-help-bounces r-project.org
>>> 03/16/201
1) In my very humble opinion R^2 can't be negative, at least for data for
which it sound to use linear model.
Or the data would have to be utterly wrong to fit them with linear model.
2) I don't want to fit data with linear model of zero intercept.
3) I dont know if I understand correctly. Im 100%
On 2011-03-16 15:02, Ben Bolker wrote:
ria.buffalo.edu> writes:
lm(y~x+0) yields the regression on x without the constant, i.e., y=bx+e,
not y = a +e
derek gmail.com>
Sent by: r-help-bounces r-project.org
03/16/2011 03:49 PM
Would someone like to (please!) write this up and submi
ria.buffalo.edu> writes:
>
> lm(y~x+0) yields the regression on x without the constant, i.e., y=bx+e,
> not y = a +e
>
> derek gmail.com>
> Sent by: r-help-bounces r-project.org
> 03/16/2011 03:49 PM
>
Would someone like to (please!) write this up and submit it to
Kurt Hornik for inclu
On Thu, Mar 17, 2011 at 10:01 AM, derek wrote:
> It states summary.lm:
>
> r.squared R^2, the ‘fraction of variance explained by the model’,
>
> R^2 = 1 - Sum(R[i]^2) / Sum((y[i]- y*)^2),
>
> where y* is the mean of y[i] if there is an intercept and zero otherwise.
>
> Why to use different f
It states summary.lm:
r.squared R^2, the ‘fraction of variance explained by the model’,
R^2 = 1 - Sum(R[i]^2) / Sum((y[i]- y*)^2),
where y* is the mean of y[i] if there is an intercept and zero otherwise.
Why to use different formula when intercept is set to zero?
I tried to compute R^2
lm(y~x+0) yields the regression on x without the constant, i.e., y=bx+e,
not y = a +e
derek
Sent by: r-help-boun...@r-project.org
03/16/2011 03:49 PM
To
r-help@r-project.org
cc
Subject
[R] Strange R squared, possible error
k=lm(y~x)
summary(k)
returns R^2=0.9994
lm(y~x) is supposed
?summary.lm
The R^2 section explains that R^2 is computed differently depending
on whether or not an intercept is in the model.
-- Bert
On Wed, Mar 16, 2011 at 12:49 PM, derek wrote:
> k=lm(y~x)
> summary(k)
> returns R^2=0.9994
>
> lm(y~x) is supposed to find coef. a anb b in y=a*x+b
>
> l=lm
Hi Derek,
R^2 doesn't mean the same thing when you omit the intercept, as has
been discussed on this list before. See
http://r.789695.n4.nabble.com/lm-without-intercept-td3312429.html
Best,
Ista
On Wed, Mar 16, 2011 at 3:49 PM, derek wrote:
> k=lm(y~x)
> summary(k)
> returns R^2=0.9994
>
> lm(y~
22 matches
Mail list logo
