4 Mx Example: Path Coefficients Model

With the introduction from the previous Sections and Chapters, we are now in a position to set up a simple genetic model using Mx. The script in Appendix  fits a simple univariate genetic model, estimating path coefficients, to covariance matrices for two like-sex twin groups: MZ twin pairs reared together, and DZ twin pairs reared together. The script is written to ignore information on means. The full path diagram is given in Figure 6.1

Figure 6.1: Univariate model for data from monozygotic (MZ) or dizygotic (DZ) twins reared together (T) or apart (A). Genetic and environmental latent variables cause the phenotypes $P_1$ and $P_2$

We have drawn this figure to correspond to the variables in the model. The latent genetic and environmental variables and $D$ cause the observed variables $P_1$ and $P_2$. The script is written to fit a model with free parameters $e, a$, and $d$, and fixing $c$ to zero -- implying that there are no effects of shared environment on BMI. The script is extensively documented using the comment facility in Mx: any line beginning with an exclamation mark is interpreted as a comment. We shall consider this first example Mx script in detail. Please note that reading this section is not a substitute for reading in detail the Mx manual [], but merely a quick introduction to the essentials of a Mx script for genetic applications.