A comparative study of the effect of the position of outliers on classical and nontraditional approaches to the two-group classification problem |
| |
| Institution: | 1. Department of Biomedical Engineering, Middle East Technical University, Ankara 06800, Turkey;2. Department of Statistics, Middle East Technical University, Ankara 06800, Turkey;1. Cranfield University, School of Energy, Environment and Agrifood, Cranfield, MK430AL, UK;2. Lancaster Environment Centre, Lancaster University, LA1 4YQ, UK;3. Engineering Innovation Centre, South University of Science and Technology of China, 1088 Xue Yuan Da Dao, Nanshan, Shenzhen, Guangdong, 518055, China;4. Department of Soil Pollution Control, Chinese Research Academy of Environmental Sciences (CRAES), 8 Dayangfang BeiYuan Road., Chaoyang District, Beijing 100012, China;5. Institute of Soil Science, Chinese Academy of Science (ISSAS), 71 East Beijing Road, Nanjing, 210008, China;6. The Institute of Urban Environment (IUE), Chinese Academy of Sciences (CAS), 1799 Jimei Road, Xiamen 361021, China;7. The Administrative Centre for China''s Agenda21 (ACCA21), 8 Yuyuantan Nanlu, Haidian District, Beijing 100038, China;8. Department of Science, Technology & Innovation, British Consulate-General Guangzhou, 5 Zhujiang Road West, Zhujiang New Town, Guangzhou 510623, China;9. Arup, Energy and Resources, 6th floor, 3 Piccadilly place, Manchester M3 1 BN, UK;10. CL:AIRE, 32 Bloomsbury Street, London WC1B 3QJ, UK;11. University of Brighton, Environment and Technology, Moulsecoomb, Brighton BN2 4GJ, UK;12. School of Geography, The University of Nottingham, University Park, Nottingham, NG7 2RD, UK & Land Quality Management Ltd, University of Innovation Park, Sir Colin Campbell Bldg, Nottingham NG7 2TU, UK;13. Department for Environment, Food and Rural Affairs (DEFRA, UK), Nobel House, 17 Smith Square, London, SW1P 3JR, UK;14. RAW, Randall and Walsh Associated Limited, 339 Yorktown road, Sandhurst GU47 0PX, UK;15. Environment Agency (England), Horizon House, Deanery Road, Bristol, BS1 5AH, UK;p. Atkins, Water Ground and Environment, Epsom, KT18 5BW, UK and Nottingham University, Ningbo, 199 Taikang E Rd, Yinzhou, Ningbo, Zhejiang 315100, China;q. British Geological Survey, Keyworth, Nottingham NG12 5GG, UK;r. Chinese Academy for Environmental Planning, 8 Dayangfang BeiYuan Road, Chaoyang District, Beijing 100012, China;s. School of Environment, Tsinghua University, Haidian, Beijing 100084, China;t. AECOM, Bridgewater House, Whitworth Street Manchester, M1 6LT, UK;u. KTN, Innovation Suite, The Heath, Runcorn, Cheshire WA7 4QX, UK;1. School of Chemistry and Environmental Engineering, Wuhan Institute of Technology, Wuhan 430205, PR China;2. State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan 430074, PR China;3. School of Xingfa Mining Engineering, Wuhan Institute of Technology, Wuhan 430205, PR China |
| |
| Abstract: | Popularity of nontraditional approaches to the statistical classification problem has resulted from the potential of these techniques to outperform the standard parametric procedures under conditions when nonnormality is present. Thus proponents of these nontraditional models have recommended these models when outliers are in the data. However, research showing that these nontraditional models' performances can vary widely depending on where the outlier data are located has not been fully illustrated. The research in this paper demonstrates how the mathematical programming approaches and the nearest neighbor discriminant models can be affected by the position of contaminated normal data and that each of the models studied in this paper may not be robust to all types of outliers in the data. The results of this paper are also important because the study compares two recently proposed mathematical programming models as well as two versions of the nearest neighbor model with the standard classical parametric models. This combination of classification models does not appear to have been studied together under conditions of contaminated normal data in which numerous positions of the outliers are considered. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
