Experimentreadr and Faq how are error bars calculated regression model: Difference between pages

From Eigenvector Research Documentation Wiki
(Difference between pages)
Jump to navigation Jump to search
imported>Jeremy
(New page: ===Purpose=== Read an Experiment file and import corresponding files. ===Synopsis=== :x = experimentreadr(''filename'') ===Description=== Experiment files include a list of data files...)
 
imported>Lyle
No edit summary
 
Line 1: Line 1:
===Purpose===
===Issue:===


Read an Experiment file and import corresponding files.
How are the error bars calculated for a regression model and can they be related to a confidence limit (confidence in the prediction)?


===Synopsis===
===Possible Solutions:===


:x = experimentreadr(''filename'')
The error bars reported for inverse least squares models (and from the [[Ils_esterror]] function) represent the estimation error for each prediction, see:


===Description===
Faber, N.M. and Bro, R., Chemomem. and Intell. Syst., 61, 133-149 (2002)


Experiment files include a list of data files and their corresponding
They can be read as a standard deviation of the estimate. However because the underlying distribution is not clearly known (and is a matter of research), a confidence limit is not reported.  
"properties of interest" (y-values). An experiment file is expected to be
a plain text file or a Microsoft Excel-formatted file. The file should
consist of at least one column of text strings indicating the files to be
read and used as samples in the X-block of a regression or classification
model and one column of numerical values indicating the values to use as
the corresponding y-block. If no experiment file filename is specified on
the command line, the user is prompted to locate a suitable file.  


Once loaded, the experiment file can be manipulated, excluding samples
using the include field of the Row Labels tab, or y-block columns using
the include field of the Column Labels tab. Samples can be marked as in
the Calibration or Validation set using the Row Labels tab.


When all manipulations are complete, the user clicks the check-mark
'''Still having problems? Please contact our helpdesk at [mailto:helpdesk@eigenvector.com helpdesk@eigenvector.com]'''
toolbar button to import all the indicated files and automatically load
the experiment data into the Analysis GUI.  


====X-block File Formats====
[[Category:FAQ]]
The x-block files named can be in any standard readable
file format. However, experiment files do not currently allow for any
multi-file formats. Named files must contain only one sample (row) of
data per file.
 
====Header Row====
An experiment file can include an optional header row for the
filenames and properties of interest. This row can contain text lables
which will be used to label the y-block columns (i.e. giving a text
description of the property of interest.)
 
====Calibration/Validation====
Experiment files can also contain information
used to split the data into calibration and validation sets. To use this
feature, include an additional column with the keywords "Calibration" or
"Validation" next to each file. When the experiment is imported, the data
will be automatically loaded into the appropriate data blocks.
NOTE: other valid synonyms include (all are case insensitive)
:Calibration = Cal = C   
:Validation  = Val = V = Test = T
 
===Example===
<tt>
  filename,concentration,cal/val
  file1.spc,13.2,cal
  file2.spc,19.0,cal
  file3.spc,5.3,cal
  file4.spc,8.3,val
</tt>
 
The above experiment file would define an experiment with three samples with X-block data
stored in the indicated files and y-values of 13.2, 19.0, 5.3, and 8.3
(with a text description of the y-values as "concentration"). The first
three files would be used for calibration, the last file for validation.
 
===See Also===
 
[[autoimport]], [[jcampreadr]], [[spcreadr]], [[xclreadr]]

Revision as of 12:07, 5 December 2018

Issue:

How are the error bars calculated for a regression model and can they be related to a confidence limit (confidence in the prediction)?

Possible Solutions:

The error bars reported for inverse least squares models (and from the Ils_esterror function) represent the estimation error for each prediction, see:

Faber, N.M. and Bro, R., Chemomem. and Intell. Syst., 61, 133-149 (2002)

They can be read as a standard deviation of the estimate. However because the underlying distribution is not clearly known (and is a matter of research), a confidence limit is not reported.


Still having problems? Please contact our helpdesk at helpdesk@eigenvector.com