Considering the Outliers in Regression Analysis

In this paper, we concentrate the issues of powerful model determination and exception recognition in linear regression.Regression analysis is the most broadly utilized strategy for fitting models to data. The consequences of information examinationin light of linear regressions are profoundly touchy to model decision and the presence of outliers in the data. To begin with, wetalk about the issue of strong model determination. Numerous techniques for performing model determination were planned withthe standard mistake model and minimum squares estimation as a top priority. At long last, we examine the issue of outlierdetection. Notwithstanding model determination, exceptions can unfavorably impact numerous different results of relapse basedinformation investigation. We portray another exception demonstrative instrument, which we call analytic information follows.This device can be utilized to identify outliers and study their impact on an assortment of relapse measurements. We exhibit ourapparatus on a few data sets, which are considered benchmarks in the field of outlier detection.

Keywords:

data mining , robust model , outlier detection , linear regression