Estimation of Principal Components Regression Coefficients

B. Saikia, R. Singh

Abstract: This paper looks into a method of principal components regression as solution of multicollinearity introducing all the indices of multicollinearity diagnosis. Using this technique, some fairly precise estimates of the coefficients are obtained. This special property of the principal components regression made it superior to the method of ordinary least squares in the presence of multicollinearity in the data. An example is utilized to describe how principal components regression analysis (including all calculating processes) becomes fruitful in case of ill-conditioned explanatory variables.

Keywords: Least Squares, Multicollinearity, Principal Components Analysis, Tolerance, Variance Inflation Factor and Principal Components Regression

How to Cite?: B. Saikia, R. Singh, "Estimation of Principal Components Regression Coefficients", Volume 3 Issue 9, September 2014, International Journal of Science and Research (IJSR), Pages: 2437-2440, https://www.ijsr.net/getabstract.php?paperid=SEP14707, DOI: https://dx.doi.org/10.21275/SEP14707