About MSC Log Out | Topics | Search
Moderators | Register | Edit Profile

NIR Discussion Forum » Bruce Campbell's List » Chemometrics » About MSC « Previous Next »

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

lee (doublel)
New member
Username: doublel

Post Number: 5
Registered: 8-2009
Posted on Wednesday, January 12, 2011 - 5:08 am:   

Dear Jim,
Thank you Jim. I have tried SNV. but the result was not better than MSC. It is strange.
best wishes.

lee
Top of pagePrevious messageNext messageBottom of page Link to this message

Jim Burger (jburger)
New member
Username: jburger

Post Number: 4
Registered: 11-2010
Posted on Wednesday, January 12, 2011 - 3:37 am:   

Dear Lee,

You might also consider trying an SNV pretreatment for comparison. There are several papers now that suggest MSC and SNV can be very very similar, however you don't provide a target spectrum for SNV.

Best wishes,
Jim
Top of pagePrevious messageNext messageBottom of page Link to this message

lee (doublel)
New member
Username: doublel

Post Number: 4
Registered: 8-2009
Posted on Wednesday, January 12, 2011 - 2:53 am:   

Dear Jose,

Thank you for your answer. When I used the MSC model created using calibration, the prediction was good. However, when I used the MSC model created with validation, the result was worse than that without pretreatment. Thank you very much.
Best regards

lee
Top of pagePrevious messageNext messageBottom of page Link to this message

Jose Miguel Hernandez Hierro (jmhhierro)
Member
Username: jmhhierro

Post Number: 14
Registered: 4-2008
Posted on Wednesday, January 12, 2011 - 2:33 am:   

Dear Lee,

I think you should use the same model.Each spectrum has been shifted and scaled using for instance the mean spectrum.You should perform the same standarization in the validation process.

Best Regards

Jose
Top of pagePrevious messageNext messageBottom of page Link to this message

lee (doublel)
New member
Username: doublel

Post Number: 3
Registered: 8-2009
Posted on Wednesday, January 12, 2011 - 1:39 am:   

hello everyone,
Recently I met a problem, when I did spectra transform with MSC using Unscrambler, the sofrware prompted me using a MSC model. My problem is which model should I select when I do validation, the one I used in calibration or created a new one?
best regards

lee

Add Your Message Here
Posting is currently disabled in this topic. Contact your discussion moderator for more information.