Skip to main content
Figure 4 | BMC Medical Research Methodology

Figure 4

From: Multi-state model for studying an intermediate event using time-dependent covariates: application to breast cancer

Figure 4

Outcome variable “death”, approach J: Plot oflog (c(d,t-d)) for two of the four models given in Table 1 and the FP approach (J.FP).The estimates for all models in Table 1 are given. The Akaike information criterion (AIC) of the best J.FP model is 8257.48 and of the model J.I, ignoring the intermediate event, 8366.40. The best FP is (t - d)-1. All coefficients are significantly (α-level 0.05) different from zero. With respect to the AIC, the J.FP model is the best one but model J.IV is almost equivalent in terms of the AIC and in terms of the shape of log (c(d,t - d)).

Back to article page