Skip to main content
. 2014 Sep 19;5:883. doi: 10.3389/fpsyg.2014.00883

Table 1.

Comparison of different process models within individuals.

Process model npars −2LogL AIC BIC χ2-increase (relative to) df p
Subject 7
VAR (0) 37 1089 1163 1255 10.377 (VAR (1)) 4 0.035
6.993 (VAR (1)*) 1 0.008
VAR (1) 41 1079 1161 1263      
VAR (1)* 38 1082 1158 1253 3.384 (VAR (1)) 3 0.336
VAR (2) 45 1095 1185 1297      
Subject 13
VAR (0) 37 1522 1596 1689 5.221 (VAR (1)) 4 0.265
VAR (1) 41 1517 1599 1702      
VAR (2) 45 1515 1605 1718      
Subject 22
VAR (0) 37 1212 1286 1378 23.655 (VAR (1)) 4 0.000
VAR (0)* 36 1214 1286 1376 1.815 (VAR (0)) 1 0.178
VAR (1) 41 1188 1270 1373      
VAR (1)* 37 1202 1276 1368 13.366 (VAR (1)) 4 0.010
VAR (2) 45 1161 1251 1363.7      
VAR (2)* 39 1189 1267 1364.1 27.390 (VAR (2)) 6 0.000

Model variants denoted with an asterisk are pruned with respect to simultaneous and lagged relationships. The relatively best fitting model according to AIC and BIC is set in italics. χ 2-differences are reported for nested models.