Confidence Interval For The Estimation of The Correlation Coefficient

Abstract

 Pearsons correlation coefficient is used to measure the influence of one quantitative variable on another quantitative variable. Basing on the sign of correlation, the type of dependence (positive/negative) can be decided in bi-variate data. The estimation of parameter can be done in two methods, point estimation and interval estimation. In this paper, various methods to find the confidence interval for the correlation are discussed. As the population correlation coefficient is estimated by the Pearsons correlation coefficient, the Monte-Carlo simulation will give the approximate the estimation of the Pearsons correlation coefficient ( r ). The Fisher Z, Bootstrap method and variance reduction methods are discussed in this paper.