Perform multiple linear regression for spectra.

Purpose
This app can be used to peform multiple linear regression of spectral data.
The key benefit is that all coeffecients are forced to be positive, and the sum of the coefficients can be constrainted to a fixed number such as 1 or 100.

Installation
Download the file "csr.opx", and then drag-and-drop onto the Origin workspace. An icon will appear in the Apps gallery window.

Operation

1. Set column designation of input data in one of the three ways: 
   • "XYYY..." if all spectra share the same X value;
   • "X1Y1, X2Y2, X3Y3..." if each spectrum has different X value or has different number of size; 
   • "YYY..." if using sampling interval for each spectrum.
2. Click the icon in the Apps Gallery window to open the dialog.
3. Specify Individual Spectra by selecting multiple Y columns.
4. Specify Composite Spectrum by selecting a single Y column.
5. In the Fit Control tab, check Fix Intercept At 0 or check Constrain Sum of Coefficients to while specifying a positive value if you want to fit with linear constraints.
6. Click Preview to view the fitted and raw composite spectrum.
7. Specify output quantities, including parameters, statistics, plots, etc.

Sample OPJU File

The App ships with a sample OPJU file. In the Apps Gallery, right-click on the Composite Spectrum Regression icon and select Show Samples Folder. From the window that opens,  drag and drop Composite Spectrum.opju  into Origin to open the project. The project includes a page of notes. Note: If saving the OPJU after changing it, it is best to save it to a different location than the one it was opened from (e.g. save to your User Files Folder).