Skip to main content
. Author manuscript; available in PMC: 2017 Mar 1.
Published in final edited form as: Psychometrika. 2014 Nov 22;81(1):102–134. doi: 10.1007/s11336-014-9431-z

Figure 3.

Figure 3

ad Plots of the RMSEs and RDSEs across the four true–fitted initial condition specifications and two sample size configurations. F–F Fixed–fixed, F–R fixed–random, R–F random–fixed, R–R random–random. The numbers in the plots indicate parameter type: 1 fixed effects dynamic parameters, including ζ0, ζ1, ζ2, μx1, and μx2, 2 measurement parameters, including the intercept parameters μ1–μ3 and factor loadings λ21 and λ31, 3 the random effect variance for the amplification parameter, σbζ2, and 4 random effect variance and covariances for the initial conditions, including σbx12,σbx22 and σbx1,x2. To avoid skewing the graphical presentation of results from the remaining conditions, the high RMSEs and RDSEs for σbζ2 (i.e., parameter type 3) in the R–F condition were omitted from the plots.