R help - Aug 2010 - Linear regression equation and coefficient matrix

Hi, 

I have 20*60 data matrix (with some NAs) and I wish to perfom a  Pearson
correlation coefficient matrix as well as simple linear regression equation
and coefficient of determination (R2) for every possible combination. Any
tip/idea/library/script how do to so. 

Thanks, 
As hz 

-- 
View this message in context:
http://r.789695.n4.nabble.com/Linear-regression-equation-and-coefficient-matrix-tp2329804p2329804.html
Sent from the R help mailing list archive at Nabble.com.

On Wed, Aug 18, 2010 at 6:09 AM, ashz <ashz at walla.co.il>
wrote:
>
> Hi,
>
> I have 20*60 data matrix (with some NAs) and I wish to perfom a ?Pearson
> correlation coefficient matrix as well as simple linear regression equation
The correlation matrix can be readily obtained by calling cor() on the
entire matrix.
> and coefficient of determination (R2) for every possible combination. Any
> tip/idea/library/script how do to so.
So you have 60 variables, and you want every possible combination?  I
may be mistaken, but isn't that 60! if you ignore any interactions?
If so, this strikes me as a job for

library(fortunes)
fortune("hundreds")

>
> Thanks,
> As hz
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/Linear-regression-equation-and-coefficient-matrix-tp2329804p2329804.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
Joshua Wiley
Ph.D. Student, Health Psychology
University of California, Los Angeles
http://www.joshuawiley.com/

Hi,

Thanks, the cor() works.

Regarding the simple linear regression equation (mainly, the slope
parameter) and r2. I think I was not writing it well. I need to do it just
for the columns. If I have a, b, c, d columns I wish to compute the relation
of there data,  e.g., between a-b, a-c, a-d, b-a, b-c, b-d, etc. 

I hope it is clear now and an help will be great.

Thanks.

-- 
View this message in context:
http://r.789695.n4.nabble.com/Linear-regression-equation-and-coefficient-matrix-tp2329804p2329948.html
Sent from the R help mailing list archive at Nabble.com.

Well you already have the correlation matrix, so you can just work from there. 
One of the profound equations in statistics is that r^2=r^2, meaning that to get
r2 (as you call it, assuming you mean the coefficient of determination) just
square the elements of the correlation matrix.

To get regression slopes, one equation is just b=s_y/s_x * r, so use the var
function to get variances, diag to pull out just the diagonal, sqrt to get the
standard deviations and then either matrix multiplication or the sweep function
to multiply (divide) the correlations by the appropriate standard deviations and
you will have a matrix of slopes.

-- 
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at imail.org
801.408.8111

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of ashz
> Sent: Wednesday, August 18, 2010 8:43 AM
> To: r-help at r-project.org
> Subject: Re: [R] Linear regression equation and coefficient matrix
> 
> 
> Hi,
> 
> Thanks, the cor() works.
> 
> Regarding the simple linear regression equation (mainly, the slope
> parameter) and r2. I think I was not writing it well. I need to do it
> just
> for the columns. If I have a, b, c, d columns I wish to compute the
> relation
> of there data,  e.g., between a-b, a-c, a-d, b-a, b-c, b-d, etc.
> 
> I hope it is clear now and an help will be great.
> 
> Thanks.
> 
> --
> View this message in context: http://r.789695.n4.nabble.com/Linear-
> regression-equation-and-coefficient-matrix-tp2329804p2329948.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.

Dear Greg,

Thanks for the tip. As I am new in R can you please provide me a script how
do to so. It will help my learning process.

Thanks,
Asher
-- 
View this message in context:
http://r.789695.n4.nabble.com/Linear-regression-equation-and-coefficient-matrix-tp2329804p2330867.html
Sent from the R help mailing list archive at Nabble.com.

Here is a quick example:


c1 <- cor(iris[,-5])
s1 <- sqrt(diag(var(iris[,-5])))

betas <- diag( s1 ) %*% c1 %*% diag( 1/s1 )

# now compare:

coef( lm( Sepal.Length ~ Sepal.Width, data=iris ) )[2]
betas[1,2]


But if you cannot work that out on your own, then you really should review
linear algebra and linear model theory.

Also when dealing with us "absent minded professor" types (I now have
adjunct status at 2 universities, can I rationalize being twice as absent minded
now?) it is best to include some context in your post (quote some of the
previous conversation) to help us remember which of many discussions we have
been part of recently you are asking about.

-- 
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at imail.org
801.408.8111

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of ashz
> Sent: Thursday, August 19, 2010 3:02 AM
> To: r-help at r-project.org
> Subject: Re: [R] Linear regression equation and coefficient matrix
> 
> 
> Dear Greg,
> 
> Thanks for the tip. As I am new in R can you please provide me a script
> how
> do to so. It will help my learning process.
> 
> Thanks,
> Asher
> --
> View this message in context: http://r.789695.n4.nabble.com/Linear-
> regression-equation-and-coefficient-matrix-tp2329804p2330867.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.