From: Thomas on
Actually I change all my analytical data for the use in matlab. This applies LA-ICP-MS data and also the IR-Spectroscopy. I read somewhere the PCA (principal componet analysis) should here a possibility also without the statistical toolbox. Here my Problem: I have actually 60 FT-IT-spectra, composed of 2 or three phases. Because I work in the mineralogical science a good database does not exist. I must do it myself. The spectra of the four components are known and they exists as numerical data. What can I do to calculate the values of my mixture with the CPA? How I should tell the MATLAB to set the mixture and the spectra of the pure phases in the correct ratio for the PCA?
We used for the analysis a THERMO-spectrometer, not a Perkin-Elmer (in the second case I can import the data directly into the matlab with a special file).

And I have a second problem, but not very difficult. I calculate my analytical data with matlab, get a lot of 2-D and 3-D-plots (ca. 15) and I want to save this. What can I do to tell the matlab to save the data (with xlswrite or save plot) with the actually name of the worksapace? It is a lot of work to change the names seperately in the M-files. The Matlab is very fast, the change of the M-file in every case for this 25 plots will not be very effective.