• Journal of the Korean Data and Information Science Society
• /
• 제17권4호
• /
• pp.1169-1180
• /
• 2006

Partial least squares regression (PLS) is a biased, non-least squares regression method and is an alternative to the ordinary least squares regression (OLS) when predictors are highly collinear or predictors outnumber observations. One way to understand the properties of biased regression methods is to know how the estimators shrink the OLS estimator. In this paper, we introduce an expression for the shrinkage factor of PLS and develop a new shrinkage expression, and then prove the equivalence of the two representations. We use two near-infrared (NIR) data sets to show general behavior of the shrinkage and in particular for what eigendirections PLS expands the OLS coefficients.

• 한국분말재료학회지
• /
• 제10권2호
• /
• pp.83-88
• /
• 2003

Nanocrystalline materials of Ni and Ni-Cu alloy have been synthesized by the pulsed wire evaporation (PWE) method and these abnormal magnetic properties in the magnetic ordered state have been characterized using both VSM and SQUID in the range of high and low magnetic fields. Ni and Ni-Cu particles with an average size of 20 to 80 nm were found to influence magnetic hysterisis behavior and the results of powder neutron diffraction patterns and saturation magnetization curves are shown to indicate the absence of the NiO phase. The shifted hysterisis loop and irreversibility of the magnetization curve in the high field region were observed in the magnetic-ordered state of both Ni and Ni-Cu. The virgin magnetization curve for Ni slightly spillover on the limited hysterisis loop ($\pm$20kOe). This irreversibility in the high field of 50 kOe can be explained by non-col-linear behavior and the existence of the metastable states of the magnetization at the surface layer (or core) of the particle in the applied magnetic field. Immiscible alloy of Cu-Ni was also found to show irreversibility having two different magnetic phases.