Pasternak, H.

Schmilovitch, Z.

Fallik, E.

Edan, Y., Dept. of Indust. Eng. and Management, Ben-Gurion University of the Negev, Beer Sheva, Israel

Schmilovitch, Z.

Fallik, E.

Edan, Y., Dept. of Indust. Eng. and Management, Ben-Gurion University of the Negev, Beer Sheva, Israel

High intercorrelation between absorbance at different wavelengths is common in near infrared analysis and was observed in an experiment to determine lycopene in tomatoes. Simulation analysis and experiments were conducted to estimate the effects of this problem on the estimators and on the predictive ability of linear regression and ridge regression. Applying linear regression to the experimental data resulted in very large parameter values, implying poor predictive ability. When linear regression gives very large parameter values, the estimated parameters are practically random numbers and are not correlated to the true ones. Ridge regression yielded estimators with normal values, but which are still poorly correlated with the true parameters. However, the predictive ability of the derived equation is good and may be used in practice to determine lycopene content in tomatoes since it is relatively easy to update.

Overcoming Multicollinearity in Near Infrared Analysis for Lycopene Content Estimation in Tomatoes by Using Ridge Regression

29

Pasternak, H.

Schmilovitch, Z.

Fallik, E.

Edan, Y., Dept. of Indust. Eng. and Management, Ben-Gurion University of the Negev, Beer Sheva, Israel

Schmilovitch, Z.

Fallik, E.

Edan, Y., Dept. of Indust. Eng. and Management, Ben-Gurion University of the Negev, Beer Sheva, Israel

Overcoming Multicollinearity in Near Infrared Analysis for Lycopene Content Estimation in Tomatoes by Using Ridge Regression

High intercorrelation between absorbance at different wavelengths is common in near infrared analysis and was observed in an experiment to determine lycopene in tomatoes. Simulation analysis and experiments were conducted to estimate the effects of this problem on the estimators and on the predictive ability of linear regression and ridge regression. Applying linear regression to the experimental data resulted in very large parameter values, implying poor predictive ability. When linear regression gives very large parameter values, the estimated parameters are practically random numbers and are not correlated to the true ones. Ridge regression yielded estimators with normal values, but which are still poorly correlated with the true parameters. However, the predictive ability of the derived equation is good and may be used in practice to determine lycopene content in tomatoes since it is relatively easy to update.

Scientific Publication

You may also be interested in