• Journal of the Korean Data and Information Science Society
• /
• v.21 no.5
• /
• pp.901-908
• /
• 2010

A geometrical description method is proposed to represent the process of the forward selection and backward elimination methods among many variable selection methods for multiple regression models. This graphical method shows the process of the forward selection and backward elimination on the first and second quadrants, respectively, of half circle with a unit radius. At each step, the SSR is represented by the norm of vector and the extra SSR or partial determinant coefficient is represented by the angle between two vectors. Some lines are dotted when the partial F test results are statistically significant, so that statistical analysis could be explored. This geometrical description can be obtained the final regression models based on the forward selection and backward elimination methods. And the goodness-of-fit for the model could be explored.

• Korean Journal of Agricultural Science
• /
• v.39 no.4
• /
• pp.603-612
• /
• 2012

In this research, the feasibility of non-destructive measurement technique of pungency measurement was investigated for the red-pepper powders produced in different domestic areas in South Korea. The near-infrared absorption spectra in the range of 1100 nm~2300 nm was used to measure capsaicinoids content in red-pepper powders by using a NIR spectroscopy equipped with Acousto-optic tunable filters (AOTF). Fourth three different red-pepper powders from 14 different locations were collected and separated in three different particle size (below 0.425 mm, 0.425~0.71 mm, 0.71~1.4 mm) for the spectral measurements. The partial least square regression (PLSR) models to predict the capsaicinoids content depends on particle size were developed with the measured spectra. The determinant coefficients and standard errors of the developed models for the red-pepper powders of below 0.425 mm, 0.425~0.71 mm, and 0.71~1.4 mm were in the range of 0.859~0.887 and 12.90~12.99 mg/100 g, respectively. The PLS model with the pretreatment of Standard Normal Variate (SNV) for the red-pepper powders below 1.4 mm particle size showed the best performance with the determinant coefficient of 0.844 and the standard error of 14.63 mg/100 g.