Browse > Article
http://dx.doi.org/10.5351/KJAS.2007.20.2.395

A Graphical Method for Evaluating the Effect of Outliers in One- and Two-Variate Data  

Jang, Dae-Heung (Division of Mathematical Sciences, Pukyoung National University)
Publication Information
The Korean Journal of Applied Statistics / v.20, no.2, 2007 , pp. 395-407 More about this Journal
Abstract
Outliers distort many measures for data analysis. We can propose dandelion seed plot as a graphical tool for evaluating the effect of outliers in one-and two-variate data. We can draw mean-variance dandelion seed plots using linked curves which are made by changing weights from 1 to 0 for each datum. Similarly we can also draw covariance-correlation-coefficient dandelion seed plots. This graphical method can be a useful tool for elementary statistics education in college.
Keywords
Mean-variance dandelion seed plot; covariance-correlation-coefficient dandelion seed plot;
Citations & Related Records
연도 인용수 순위
  • Reference
1 渡部洋, 鈴木規夫, 山田文康, 大塚雄作(1988). 探索的テ-タ解析入門, 朝倉書店
2 Cook, R. D.(1986). Assessment of local influence, Journal of Royal Statistical Society, Ser. B, 48, 133-169
3 Hampel, F. R.(1974). The influence curve and its role in robustness, The Annal of Statistics, 45, 383-393
4 Maronna, R. A., Martin, D. and Yohai, V. J. (2006). Robust Statistics, John Wiley & Sons, New York