R help - Mar 2009 - PLS regression on near infrared (NIR) spectra data

Dear collegues,

I´ ve worked with near infrared (NIR) spectroscopy to assess chemical,
physical, mechanical and anatomical properties of wood.

I use "The Unscrambler" software to correlate the matrix of dependent
variables (Y) with the matrix of spectral data (X) and I would like to
migrate to R. The matrix of spectral variables is very large (2345 columns
and n lines, where n = samples), so we used Partial Least Squares Regression
to predict a variable y (content of cellulose, for instance) based on the
spectral variables, which are the NIR wavelengths.

I am new here (since jan2009) and up to now, I not seen anyone commenting
about principal component analysis and regression PLS to analyze spectral
information in R system. Sorry, I am a R starter...

Anybody have any package, or trick to suggest me?

Grateful for yours information!
 --
Paulo Ricardo Gherardi Hein
PhD candidate at University of Montpellier 2
CIRAD - PERSYST Department
Research unit: Production and Processing of Tropical Woods - TA B-40/16
73 rue Jean-François Breton 34398 Montpellier Cedex 5, France
phone: +33 4 67 61 44 51
skype: paulo_hein
email: paulo.hein@cirad.fr

	[[alternative HTML version deleted]]

Hi, take a look on pls package and it's documentation, there are
examples also for NIR data.

http://mevik.net/work/software/pls.html

Article form "Journal of Statistical Software"

http://www.jstatsoft.org/v18/i02

Also "Caret" package can be used to evaluate pls and other regreesion
models:

http://caret.r-forge.r-project.org/Classification_and_Regression_Training.html

Best regards,

Andris

On Tue, Mar 3, 2009 at 4:49 PM, Paulo Ricardo Gherardi Hein
<phein1980 at gmail.com> wrote:
> Dear collegues,
>
> I? ve worked with near infrared (NIR) spectroscopy to assess chemical,
> physical, mechanical and anatomical properties of wood.
>
> I use "The Unscrambler" software to correlate the matrix of
dependent
> variables (Y) with the matrix of spectral data (X) and I would like to
> migrate to R. The matrix of spectral variables is very large (2345 columns
> and n lines, where n = samples), so we used Partial Least Squares
Regression
> to predict a variable y (content of cellulose, for instance) based on the
> spectral variables, which are the NIR wavelengths.
>
> I am new here (since jan2009) and up to now, I not seen anyone commenting
> about principal component analysis and regression PLS to analyze spectral
> information in R system. Sorry, I am a R starter...
>
> Anybody have any package, or trick to suggest me?
>
> Grateful for yours information!
> ?--
> Paulo Ricardo Gherardi Hein
> PhD candidate at University of Montpellier 2
> CIRAD - PERSYST Department
> Research unit: Production and Processing of Tropical Woods - TA B-40/16
> 73 rue Jean-Fran?ois Breton 34398 Montpellier Cedex 5, France
> phone: +33 4 67 61 44 51
> skype: paulo_hein
> email: paulo.hein at cirad.fr
>
> ? ? ? ?[[alternative HTML version deleted]]
>
>
> ______________________________________________
> 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.
>
>

Paulo Ricardo Gherardi Hein <phein1980 at gmail.com> writes:
> I am new here (since jan2009) and up to now, I not seen anyone commenting
> about principal component analysis and regression PLS to analyze spectral
> information in R system. Sorry, I am a R starter...
>
> Anybody have any package, or trick to suggest me?
There is the package 'pls', with Principal Component Regression (PCR)
and
Partial Least Squares Regression (PLSR).  It also contains a couple of
plots that are useful for princomp() or prcomp() analyses (PCA).

-- 
Bj?rn-Helge Mevik

Hi Paulo,

You might also want to look at something like the glmnet package (Friedman,
Hastie, and Tibshirani). This carries out penalized regression, is designed
to work with high numbers of predictors/inputs/columns and relatively few
samples/obervations/rows, and is very fast.

See: http://www-stat.stanford.edu/~hastie/Papers/glmnet.pdf

HTH, Mark.



Paulo Ricardo Gherardi Hein wrote:
> 
> Dear collegues,
> 
> I? ve worked with near infrared (NIR) spectroscopy to assess chemical,
> physical, mechanical and anatomical properties of wood.
> 
> I use "The Unscrambler" software to correlate the matrix of
dependent
> variables (Y) with the matrix of spectral data (X) and I would like to
> migrate to R. The matrix of spectral variables is very large (2345 columns
> and n lines, where n = samples), so we used Partial Least Squares
> Regression
> to predict a variable y (content of cellulose, for instance) based on the
> spectral variables, which are the NIR wavelengths.
> 
> I am new here (since jan2009) and up to now, I not seen anyone commenting
> about principal component analysis and regression PLS to analyze spectral
> information in R system. Sorry, I am a R starter...
> 
> Anybody have any package, or trick to suggest me?
> 
> Grateful for yours information!
>  --
> Paulo Ricardo Gherardi Hein
> PhD candidate at University of Montpellier 2
> CIRAD - PERSYST Department
> Research unit: Production and Processing of Tropical Woods - TA B-40/16
> 73 rue Jean-Fran?ois Breton 34398 Montpellier Cedex 5, France
> phone: +33 4 67 61 44 51
> skype: paulo_hein
> email: paulo.hein at cirad.fr
> 
> 	[[alternative HTML version deleted]]
> 
> 
> ______________________________________________
> 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.
> 
> 
-- 
View this message in context:
http://www.nabble.com/PLS-regression-on-near-infrared-%28NIR%29-spectra-data-tp22311467p22328510.html
Sent from the R help mailing list archive at Nabble.com.