Abstract
Closure is caused by normalization of a data set, and it affects any multivariate analytical method applied to that data set. Two common methods of normalizing infrared spectra (IR), to unit maximum absorbance and to unit vector length, are evaluated by measurement of library search performance. Search performance is evaluated, by the use of the Quantitative Reliability Metric (QRM), as a function of noise frequency and noise level.
Get full access to this article
View all access options for this article.
References
1.
Johansson
E.
Wold
S.
, and
Sjodin
K.
, Anal. Chem. 56 , 1685 (1984 ).
2.
Karrer
L. M.
Gordon
H. L.
Rothstein
S. M.
Miller
J. M.
, and
Jones
T. R. B.
, Anal. Chem. 55 , 1723 (1983 ).
3.
Lowery
S. R.
and
Huppler
D. A.
, Anal. Chem. 53 , 889 (1981 ).
4.
Harrington
P. B.
and
Isenhour
T. L.
. Presented at the 1st Hidden Peak Conference , Snowbird, Utah (1986 ).
5.
Harrington
P. B.
and
Isenhour
T. L.
, Anal. Chim. Acta , submitted.
6.
Grotch
S. L.
, Anal. Chem. 43 , 1362 (1971 ).
7.
Rasmussen
G. T.
and
Isenhour
T. L.
, J. Chem. Inf. Comput. Sci. 43 , 1382 (1979 ).
8.
Delaney
M. F.
Warren
F. V.
Jr.
, and
Hallowell
J. R.
Jr.
, Anal. Chem. 55 , 1925 (1983 ).
9.
Hamming
R. W.
, Digital Filters (Prentice-Hall , New Jersey , 1983 ), p. 64 .
10.
Olkin
I.
Gleser
L. J.
, and
Derman
C.
, Probability Models and Applications (Macmillan Publishing Co. , New York , 1980 ), p. 151 .