Skip to main content
. 2021 Mar 10;159:107217. doi: 10.1016/j.csda.2021.107217

Table 4.

Percentage of clustering object errors (OE) based on 1000 simulation replications when data are generated from (27) with respect to k-means (K), convex clustering (Convex), and k-means++ (KPP), directly on regression coefficients and our BIC selector in the generalized k-means.

σ c k n0=50
n0=100
K Convex KPP BIC K Convex KPP BIC
0.1 10 2 0.4 0.1 0.6 0.0 0.2 0.0 0.3 0.0
3 1.4 0.0 0.2 0.0 1.5 0.0 0.1 0.0
20 2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
3 1.5 0.0 0.0 0.0 1.5 0.0 0.0 0.0
0.2 10 2 0.3 0.1 0.5 0.0 0.2 0.0 0.3 0.0
3 14.0 12.4 16.1 0.0 1.4 0.0 0.1 0.0
20 2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
3 17.5 17.6 20.6 0.0 1.5 0.0 0.0 0.0
0.5 10 2 10.8 10.3 10.4 0.1 0.9 0.6 0.9 0.0
3 33.1 32.6 32.2 2.5 26.0 26.8 25.7 0.1
20 2 8.4 8.2 8.3 0.1 0.5 0.5 0.5 0.0
3 31.4 31.3 31.2 1.5 25.9 26.3 25.6 0.1
1.0 10 2 41.1 40.2 40.0 8.3 22.8 21.7 20.5 1.2
3 46.8 45.7 46.4 22.6 38.9 38.1 38.0 11.1
20 2 39.4 38.9 38.1 5.3 18.6 18.0 17.7 1.2
3 45.9 45.2 45.1 18.0 36.4 36.1 35.8 4.5