Skip to main content
. Author manuscript; available in PMC: 2012 Feb 23.
Published in final edited form as: J Am Stat Assoc. 2011 Sep 1;106(495):904–913. doi: 10.1198/jasa.2010.tm10053

Table 1.

Comparison of MIP and relaxed versions of the ACI. For each data set the table was constructed using 1000 training sets each with 1000 bootstrap iterations for a total of 1,000,000 computations of the optimization problem given in (12)

Data Set Coverage Difference in width Computation time
Relaxed MIP p.99 p.95 p.75 p.5 Relaxed MIP

ThreePt .948 .948 .0334 0.00 0.00 0.00 .734 3.11
Magic .944 .945 .0334 .0334 0.00 0.00 1.24 1.94
Mam. .957 .958 .0334 0.00 0.00 0.00 .904 1.88
Ion. .954 .954 .0334 0.00 0.00 0.00 1.33 3.06
Donut. .967 .968 .0667 .0334 .0334 0.00 .917 2.94
Bal. .969 .969 0.00 0.00 0.00 0.00 .977 1.69
Liver .956 .956 .0333 .0333 0.00 0.00 1.61 2.50
Spam .984 .987 .0333 .0333 0.00 0.00 1.54 3.01
Quad .959 .962 .0333 0.00 0.00 0.00 .983 1.37
Heart .960 .961 .0333 0.00 0.00 0.00 1.06 3.27