Detection of Outliers in Regression Model for Medical Data

Abstract

In regression analysis, an outlier is an observation for which the residual is large in magnitude compared to other observations in the data set. The detection of outliers and influential points is an important step of the regression analysis. Outlier detection methods have been used to detect and remove anomalous values from data. In this paper, we detect the presence of outliers in simple linear regression models for medical data set. Chatterjee and Hadi mentioned that the ordinary residuals are not appropriate for diagnostic purposes; a transformed version of them is preferable. First, we investigate the presence of outliers based on existing procedures of residuals and standardized residuals. Next, we have used the new approach of standardized scores for detecting outliers without the use of predicted values. The performance of the new approach was verifed with the real-life data.

Authors and Affiliations

Stephen Raj S| Research Scholar, Department of Statistics, Manonmaniam Sundaranar University, Abhishekapatti, Tirunelveli, Tamil Nadu, India Corresponding e-mail: stephenstats17@gmail.com, Senthamarai Kannan K| Professor, Department of Statistics, Manonmaniam Sundaranar University, Abhishekapatti, Tirunelveli, Tamil Nadu, India

Keywords

Related Articles

Effects of education on exercise (physical activity) performance of pregnant women

Pregnancy is an important period of life that may put them at risk of decreased physical activity and a large percentage of women have little knowledge about exercise during pregnancy, and their performance is not good....

Training emotional-behavioral theory of mind on reducing the problems of pre-school learners

In order to evaluate the effectiveness of teaching theory of mind reducing the emotional and behavioral problems in preschool learners 40 are selected and randomly were assigned to two experimental and control groups. A...

Evaluation tissue dissolution property of 2.5 % Sodium Hypochlorite Prepared by Hydrochloric Acid and Sodium Bicarbonate: An in vitro

Successful endodontic treatment requires chemical preparation in addition to mechanical preparation. The most common material for chemical preparations is sodium hypochlorite. One way to reduce the effects of pH adjustm...

Silhouette of substance abuse amongst an adolescent sample group from urban slums of Guwahati metro, North East India

Globally 320 million young people between ages 15-29 years are dying from alcohol related causes. Child line foundation survey report showed that 13.1% of the people involved in drug and substance abuse in India, are a...

LIMB SALVAGE IN RECURRENT GIANT CELL TUMOUR PROXIMAL END RADIUS

An 18years old male patient had undergone surgery for a giant cell tumour upper end of the right radius in Jan 2007. The tumour recurred within two months. Patient took alternative methods of treatment including indige...

Download PDF file
  • EP ID EP12408
  • DOI -
  • Views 304
  • Downloads 20

How To Cite

Stephen Raj S, Senthamarai Kannan K (2017). Detection of Outliers in Regression Model for Medical Data. International Journal of Medical Research & Health Sciences (IJMRHS), 6(7), 50-56. https://europub.co.uk/articles/-A-12408