Partial Least Squares Regression versus Multiple Linear Regression

NIR news ◽  
2011 ◽  
Vol 22 (4) ◽  
pp. 15-16 ◽  
Author(s):  
Tom Fearn
Keyword(s):  
Linear Regression ◽  
Multiple Linear Regression ◽  
Least Squares ◽  
Partial Least Squares ◽  
Partial Least Squares Regression ◽  
Least Squares Regression

scholarly journals Prediction of Lard in Palm Olein Oil Using Simple Linear Regression (SLR), Multiple Linear Regression (MLR), and Partial Least Squares Regression (PLSR) Based on Fourier-Transform Infrared (FTIR)

2018 ◽  
Vol 2018 ◽  
pp. 1-8 ◽  
Author(s):  
Siong Fong Sim ◽  
Min Xuan Laura Chai ◽  
Amelia Laccy Jeffrey Kimura
Keyword(s):  
Fourier Transform ◽  
Linear Regression ◽  
Multiple Linear Regression ◽  
Least Squares ◽  
Partial Least Squares ◽  
Partial Least Squares Regression ◽  
Palm Olein ◽  
Least Squares Regression ◽  
Simple Linear Regression ◽  
Olein Oil

Fourier-transform infrared (FTIR) offers the advantages of rapid analysis with minimal sample preparation. FTIR in combination with multivariate approach, particularly partial least squares regression (PLSR), has been widely used for adulterant analysis. Limited study has been done to compare PLSR with other regression strategies. In this paper, we apply simple linear regression (SLR), multiple linear regression (MLR), and PLSR for prediction of lard in palm olein oil. Pure palm olein oil was adulterated with lard at different concentrations and subjected to analysis with FTIR. The marker bands distinguishing lard and palm olein oil were determined using Fisher’s weights. The marker regions were then subjected to regression analysis with the models verified based on 100 training/test sets. The prediction performance was measured based on the percentage root mean square error (%RMSE). The absorption bands at 3006 cm−1, 2852 cm−1, 1117 cm−1, 1236 cm−1, and 1159 cm−1 were identified as the marker bands. The bands at 3006 and 1117 cm−1 were found with satisfactory predictive ability, with PLSR demonstrating better prediction yielding %RMSE of 16.03 and 13.26%, respectively.

scholarly journals Análisis de métodos de validación cruzada para la obtención robusta de parámetros biofísicos

2015 ◽  
pp. 55 ◽  
Author(s):  
Ll. Pérez-Planells ◽  
J. Delegido ◽  
J. P. Rivera-Caicedo ◽  
J. Verrelst
Keyword(s):  
Linear Regression ◽  
Gaussian Process ◽  
Least Squares ◽  
Partial Least Squares ◽  
Ridge Regression ◽  
Partial Least Squares Regression ◽  
Gaussian Process Regression ◽  
Least Squares Regression ◽  
Kernel Ridge Regression

<p class="Bodytext">Los métodos de regresión no paramétricos son una gran herramienta estadística para obtener parámetros biofísicos a partir de medidas realizadas mediante teledetección. Pero los resultados obtenidos se pueden ver afectados por los datos utilizados en la fase de entrenamiento del modelo. Para asegurarse de que los modelos son robustos, se hace uso de varias técnicas de validación cruzada. Estas técnicas permiten evaluar el modelo con subconjuntos de la base de datos de campo. Aquí, se evalúan dos tipos de validación cruzada en el desarrollo de modelos de regresión no paramétricos: hold-out y k-fold. Los métodos de regresión lineal seleccionados fueron: Linear Regression (LR) y Partial Least Squares Regression (PLSR). Y los métodos no lineales: Kernel Ridge Regression (KRR) y Gaussian Process Regression (GPR). Los resultados de la validación cruzada mostraron que LR ofrece los resultados más inestables, mientras KRR y GPR llevan a resultados más robustos. Este trabajo recomienda utilizar algoritmos de regresión no lineales (como KRR o GPR) combinando con la validación cruzada k-fold con un valor de k igual a 10 para hacer la estimación de una manera robusta.</p>

Research on Several Problems in Partial Least Squares Regression Analysis

2014 ◽  
Vol 8 (1) ◽  
pp. 754-758 ◽  
Author(s):  
Ju Wu
Keyword(s):  
Mathematical Model ◽  
Regression Analysis ◽  
Linear Regression ◽  
Least Squares ◽  
Partial Least Squares ◽  
Partial Least Squares Regression ◽  
Linear Regression Analysis ◽  
Multiple Linear Regression Analysis ◽  
Least Squares Regression ◽  
Analysis Method

Purpose: preliminary discussion on model prediction precision in the partial least squares regression analysis method; Method: introduce current development conditions of partial least squares regression analysis, analyze problems of traditional regression analysis method such as multiple linear regression analysis, introduce the mathematic principle and modeling method of the partial least squares regression analysis method, and conduct detailed analysis on the partial least squares regression analysis modeling and prediction by using the classical Linnerud data. Result: The partial least squares regression analysis has the basic features of the multiple linear regression analysis and principal component analysis, can precisely predict multiple data and establish a precise mathematical model; Conclusion: The partial least squares regression analysis can provide precise mathematical model and can reserve the explaining variants remarkably associated to explained variants to most extent, so it is feasible to some extent and can meet the general requirements of engineering, economy, biology and medical statistical analysis.

Sign in / Sign up

Export Citation Format

Share Document