BOOTSTRAPPING IN DETERMINATION OF RIDGE PARAMETER
Journal: International Journal of Applied Mathematics & Statistical Sciences (IJAMSS) (Vol.4, No. 6)Publication Date: 2015-11-30
Authors : OKEKE EVELYN NKIRUKA; OKEKE JOSEPH UCHENNA;
Page : 25-34
Keywords : Agriculture Finance; Credit Policy; Poverty; Rural Development;
Abstract
This article studied the application of ridge regression on multicollinear data whose ridge parameter was determined using bootstrap samples. Mean squared error of the samples and some arbitrary values were used to determine the ridge parameter that will give the minimum residual. The result of the study revealed that both the mean squared error and the smallest eigenvalue of the predictor variables of the original data play vital role in determining the ridge parameter of ridge regression.
Other Latest Articles
- Syncretic transcendence of the axiosphere of culture
- Reason and madness in the structure of the Cartesian cogito. Three interpretations of Rene Descartes' Meditation
- On philosophical and culturological views of V.V. Zenkovsky
- Astral motivs and images in the novel of Alexei Ivanov “The geographer drank his globe away”
- The categories of the lyrical character and the material world in the Joseph Brodsky’s love poetry (the collection “The new stanzas to Augusta”): the interaction features
Last modified: 2015-10-09 17:37:50