Gülhan TEMEL OREKİCİ, Semra ERDOĞAN, İrem Ersöz KAYA, Emine Arzu KANIK

Affected states soft ındependent modeling by class analogy from the relation between ındependent variables, number of ındependent variables and sample size

Objective: The aim of study is to introduce method of Soft Independent Modeling of Class Analogy (SIMCA), and to express whether the method is affected from the number of independent variables, the relationship between variables and sample size. Study Design: Simulation study. Material and Methods: SIMCA model is performed in two stages. In order to determine whether the method is influenced by the number of independ- ent variables, the relationship between variables and sample size, simulations were done. Conditions in which sample sizes in both groups are equal, and where there are 30, 100 and 1000 samples; where the number of variables is 2, 3, 5, 10, 50 and 100; moreover where the relationship between variables are quite high, in medium level and quite low were mentioned. Results: Average classification accuracy of simulation results which were carried out 1000 times for each possible condition of trial plan were given as tables. Conclusion: It is seen that diagnostic accuracy results increase as the number of independent variables increase. SIMCA method is a method in which the relationship between variables are quite high, the number of independent variables are many in number and where there are outlier values in the data that can be used in conditions having outlier values.

PDF

___

1. Srivastava MS. Methods of Multivariate Statistics. Eds:Balding DJ, Bloomfield P, Cressie NAC, A John Wiley&Sons Inc, Canada, 2002. p.246-65.
2. Mcclave JT, Benson GP, Sincich T. Statistics for Business and Eco- nomics. 7 th ed, Upper Saddle River N. J, Prentice Hall, 1998. p.551-552.
3. Lattin MJ, Carroll DJ, Green P. Analyzing Multivariate Data. Brooks/Cole-Thompson, Pacific Grove CA, 2003. p.426-428.
4. Wold S. Pattern Recognition by Means of Disjoint Principal Com- ponents Models. Pattern Recogn 1976;8:127-39. [CrossRef]
5. Sørensen B, Falk ES, Wisløff-Nilsen E, Bjorvatn B, Kristiansen BE. Multivariate analysis of Neisseria DNA restriction endonuclease patterns. J Gen Microbiol 1985;131:3099-104.
6. Bylesjö M, Rantalainen M, Cloarec O, Nicholoson JK, Holmes E, Trygg J. OPLS Discriminant Analysis:Combining the Strengths of PLS-DA and SIMCA Classification. J Chemomet 2006;20:341-51. [CrossRef]
7. Lopez-de-Alba P, Lopez-Martinez L, Cerda V, Amador-Hernandez A. Simulaneous Determination and Classification of Riboflavini Thiamine, Niotinamide and Pyridoxine in Phamaceutical Formu- lations, by UV-Visible Spectrophotometry and Multivariate Analy- sis. J Braz Chem Soc 2006;17:715-22. [CrossRef]
8. Maesschalck RD, Candolfi A, Massart DL, Heuerding S. Decision criteria for soft independent modelling of class analogy applied to near infrared data. Chemometr Intell Lab 1999;47:65-77. [CrossRef]
9. Branden KV, Hubert M. Robust classification in High Dimensions based on the SIMCA Method. Chemometr Intell Lab 2005;79:10- 21. [CrossRef]
10. Gemperline P, Webber LD. Raw materials testing using soft in- dependent modelling of class analogy analysis of near-infrared reflectance spectra. J Am Chem Soc 1989;61:138-44.
11. Esbensen KH. SIMCA:An Introduction to Classification. Hou- moller LP, eds. Multivariate data analysis in practice:An Introduc- tion to multivariate Stata Analysis and Experimental Design. 5 th ed, Camo process AS, 2005.p:348-51.
12. Dunn WJ, Emery SL, Glen WG. Preprocessing, variable selec- tion and classification rules in the application of simca pattern recognition to mass spectral data. Environ Sci and Technol 1989;23:1499-505. [CrossRef]
13. Zhu M, Shi Y, Li A, He J. A dinamic committee scheme on multi- ple-criteria linear programming classification method. Computa- tional Science ICCS. 2007;4488:401-8.
14. Chen LF, Liao HYM, Ko MT, Lin JC, Yu GY. A new LDA-based face recognition system which can solve the small sample size prob- lem. Pattern Recogn 2000;33:1713-26. [CrossRef]
15. Breiman L, Friedman JH, Olshen RA, Stone CJ. Introduction to Tree Classification. Classification and Regression Trees. 1st ed, London, Chapman & Hall, 2003. p. 18-55.
16. Friedman JH. Multivariate Adaptive Regression Splines. Ann Stat 1991;19:1-141. [CrossRef]