dataset=nasty begin description Artificial data contributed by Douglas M. Hawkins. There are at least two groups of data. I am guessing that the outliers are cases 2, 6, 10, 14, 18, 22, 26, and 30. That is, every 4 th observation starting with 2. end description begin variables col 0 = case col 1 = x1 col 2 = x2 col 3 = x3 col 4 = x4 col 5 = y end variables begin data 1 -1.2067 -1.2943 1.2127 1.2819 3.4232 2 -1.6638 -1.6529 -1.6623 -1.6809 -4.9949 3 -0.2427 0.0206 0.2449 -0.0823 0.4115 4 0.6919 -1.1272 1.4329 -1.0012 -1.0821 5 1.0816 -0.5245 0.4564 -1.038 -2.2005 6 1.3712 1.3858 1.3779 1.3959 4.1482 7 0.1056 -1.9662 1.251 0.6264 1.0322 8 0.8032 -0.3207 -1.1218 0.6604 -1.1425 9 0.9762 -1.2155 -0.6139 0.8358 -0.9901 10 -0.6182 -0.6373 -0.6532 -0.6444 -1.9149 11 0.1854 0.2714 -1.9477 1.4711 -0.0259 12 0.1863 0.0242 0.8027 -1.0168 -0.7228 13 0.2774 0.7022 0.1945 -1.1699 -1.3077 14 -0.8461 -0.8349 -0.8277 -0.8469 -2.5167 15 0.0398 1.5352 -0.8902 -0.6929 -1.1018 16 -0.4306 1.8651 -0.6333 -0.7496 -0.3752 17 0.6682 -0.5084 0.0682 -0.2101 -1.0882 18 0.647 0.6692 0.6466 0.6626 1.969 19 -2.1892 0.0681 1.3497 0.8275 4.4979 20 0.0828 0.7766 0.1442 -0.9081 -0.8214 21 0.4046 -0.3402 -0.7147 0.6673 -0.39 22 0.1049 0.0964 0.1136 0.1063 0.316 23 1.4467 -0.1116 -0.6042 -0.7499 -2.9959 24 -1.1789 1.2047 -1.8836 1.8295 2.2637 25 -1.1164 1.3412 -0.0536 -0.2664 1.3076 26 -0.2107 -0.2244 -0.2006 -0.2146 -0.6377 27 1.8686 -0.3153 -0.1768 -1.3333 -3.8837 28 -0.4579 1.4008 0.272 -1.1484 -0.1886 29 -1.3907 0.0582 0.9717 0.2856 2.7179 30 1.2157 1.1981 1.2057 1.2219 3.631 31 -1.3604 -0.1684 1.4017 0.1312 2.7614 32 0.7553 -1.3759 -1.1628 1.7504 -0.1668