Skip to main content
. 2004 Sep 14;1:8. doi: 10.1186/1742-4682-1-8

Table 5.

Results for piecewise linear regression

Interval 1 Interval 2 Interval 3
a10 0.1315 -0.0419 0.0000
a11 -42.3980 -14.1738 -14.5490
a12 0.0000 -0.8010 -0.0464
a13 8.9105 7.3653 7.6299
a14 12.7757 -0.3340 -0.1386
a15 -3.3476 -6.9121 -7.2940
a20 0.0567 -0.0197 0.0000
a21 -1.1939 14.4913 14.6792
a22 -32.3300 -14.5116 -14.6784
a23 0.6133 0.0057 -0.0205
a24 7.0917 0.1016 -0.0018
a25 7.9313 -0.1047 0.0067
a30 -0.7858 -0.0181 0.0000
a31 -130.3724 -0.2358 0.0021
a32 0.0000 0.3616 -0.0007
a33 -20.7724 -27.6129 -27.2551
a34 62.1525 0.3496 -0.0027
a35 19.1470 -0.1984 0.0006
a40 0.3164 -0.0709 0.0000
a41 -13.6819 1.1412 -0.0115
a42 0.0000 -2.1478 0.0015
a43 19.8295 18.8534 18.6927
a44 -13.3654 -19.5811 -18.5494
a45 -7.2135 -8.0985 -9.2792
a50 0.1617 -0.0393 0.0000
a51 -149.5199 -0.8195 0.0250
a52 -160.3341 0.8175 -0.0074
a53 5.7537 0.0580 -0.0304
a54 85.3050 19.0394 18.5356
a55 53.9745 -19.1183 -18.5623

The complete dataset is divided into three subsets for each variable, where the first and second extreme values serve as breakpoints. The datasets for the regression consisted of 401 data points in the interval [0,4] and resulted from a simulation in which X3 was perturbed at t = 0 to a value 5% above its steady-state value.