Classification-Based Ridge Estimation Techniques of Alkhamisi Methods

Lukman, A. F. and Ayinde, K. and Oluyemi, O. A. and Akanbi, O. B. and Onate, C.A (2018) Classification-Based Ridge Estimation Techniques of Alkhamisi Methods. Journal of Probability and Statistical Sciences, 16 (2). pp. 165-181. ISSN 1726-3328

[img] Text
JPSS32pp165-181.pdf - Published Version

Download (266kB)


Following Lukman and Ayinde [9]: review and classification of methods of estimating ridge parameters into different forms and various types, this study proposed some new ridge parameter estimation using the idea of Alkhamisi et al. [1]. The performance of the techniques was evaluated by conducting Monte-Carlo experi- ments under certain conditions and compared using relative efficiency. Results show that increase in the strength of multicollinearity resulted in increase in mean square error (MSE), which decreases as the sample size increases. Furthermore, the most preferred technique is generally in the different forms in the original and square root types. Moreover, Fixed Maximum Original (FMO) for Alkhamisi et al. [1], the proposed Varying Maximum Original (VMO) for AL4, VMO for AL6 and Harmonic Mean Original (HMO) for AL5 competes favorably. Keywords Mean square error; Monte-Carlo experiment;Ridge parameter; Relative efficiency.

Item Type: Article
Subjects: H Social Sciences > HA Statistics
Date Deposited: 20 Sep 2019 12:05
Last Modified: 20 Sep 2019 12:05

Actions (login required)

View Item View Item