Published on *OpenMx* (http://openmx.psyc.virginia.edu)

By *ferranc*

Created *07/25/2010 - 07:31*

Sun, 07/25/2010 - 07:31 — ferranc [1]

I'm learning SEM with the OpenMx documentation. First of all thanks for this I would agree your effort and attention.

I'm a beginner in SEM and I'm not sure about this question. I see a possible bug in documentation http://openmx.psyc.virginia.edu/docs/OpenMx/latest/Regression_Path.html [2]. The specification of a multiple regression and multivariate regression talk about the covariance between x and z in a y equation (resp. z equation). This relation is specified in scripts like:

# covariance of x and z

mxPath(

from="x",

to="y",

arrows=2,

free=TRUE,

values=0.5,

labels="covxz"

),

Is this correct or have to be:

# covariance of x and z

mxPath(

from="x",

to="z",

arrows=2,

free=TRUE,

values=0.5,

labels="covxz"

),

I see the same thing in the graph specification (with sigmaxy).

thanks in advance,

Ferran Carrascosa

**Links:**

[1] http://openmx.psyc.virginia.edu/users/ferranc

[2] http://openmx.psyc.virginia.edu/docs/OpenMx/latest/Regression_Path.html

[3] http://openmx.psyc.virginia.edu/thread/614

[4] http://openmx.psyc.virginia.edu/thread/582

[5] http://openmx.psyc.virginia.edu/forums/openmx-help/openmx-structural-equation-modeling