The mineral particles are classified in different textural classes according to their size. Reflectance spectrometry and reflectance spectra can be valid instruments to classify soil according to their texture. This is possible using different statistical methods, as for example, discriminant analysis. However, others multivariate methods, like multinomial logistic regression can be used, but the presence of multicollinearity between explicative variables could affects the estimation of the parameters. To solve this problem we propose an alternative way to apply the multinomial logit model and we compare its performance to the classical method and to discriminant analysis results.
A statistical model to predict soil texture through laboratory spectrometry
CAMMINATIELLO, Ida;
2010
Abstract
The mineral particles are classified in different textural classes according to their size. Reflectance spectrometry and reflectance spectra can be valid instruments to classify soil according to their texture. This is possible using different statistical methods, as for example, discriminant analysis. However, others multivariate methods, like multinomial logistic regression can be used, but the presence of multicollinearity between explicative variables could affects the estimation of the parameters. To solve this problem we propose an alternative way to apply the multinomial logit model and we compare its performance to the classical method and to discriminant analysis results.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
