beautypg.com

Regression statistics, Coefficients standard error – Ocean Optics NIR User Manual

Page 23

background image

A: Calibrating the Wavelength of the NIR Spectrometer

Independent

Dependent

Values computed from

Variable

Variables

the regression output


True Wavelength (nm)

Pixel #

Pixel #

2

Pixel

#

3

Predicted

Wavelength

Difference

253.65
296.73
302.15
313.16
334.15
365.02
404.66
407.78
435.84
546.07
576.96
579.07
696.54
706.72
727.29
738.40
751.47

175
296
312
342
402
490
604
613
694

1022
1116
1122
1491
1523
1590
1627
1669

30625
87616
97344

116964
161604
240100
364816
375769
481636

1044484
1245456
1258884
2223081
2319529
2528100
2647129
2785561

5359375

25934336
30371328
40001688
64964808

117649000
220348864
230346397
334255384

1067462648
1389928896
1412467848
3314613771
3532642667
4019679000
4306878883
4649101309

253.56
296.72
302.40
313.02
334.19
365.05
404.67
407.78
435.65
546.13
577.05
579.01
696.70
706.62
727.24
738.53
751.27

0.09
0.01

-0.25

0.13

-0.05
-0.04
-0.01

0.00
0.19

-0.06
-0.09

0.06

-0.15

0.10
0.06

-0.13

0.19

5. Use your spreadsheet or calculator to calculate the wavelength calibration coefficients. In your

spreadsheet program, find the functions to perform linear regressions.

If using Quattro Pro, look under Tools | Advanced Math

If using Excel, look under Analysis ToolPak

6. Select the true wavelength as the dependent variable (Y). Select the pixel number, pixel number

squared, and the pixel number cubed as the independent variables (X). After you execute the
regression, an output similar to the one shown below is obtained.

Regression Statistics

Multiple R

0.999999831

R Square

0.999999663

R Squared

Adjusted R Square

0.999999607

Standard Error

0.125540214

Observations 22

intercept

Coefficients Standard Error

Intercept 190.473993

0.369047536

first coefficient

X Variable 1

0.36263983

0.001684745

X Variable 2

-1.174416E-05

8.35279E-07

X Variable 3

-2.523787E-09

2.656608E-10

second coefficient

third coefficient

The numbers of importance are indicated in the above figure.

197-00000-512-02-0707

17