quiet failure without error message

Posted on
No user picture. deb193 Joined: 12/01/2010
I am not getting an error message, but the model seems to be not running. I run the ful model and get a GREEN status, then I set the paths moderatign the loadings to 0 and run the nested model ... and nothing. It even reports 0 observed statistics.

How can I troubleshoot when there is no eror message?

------------------------------------------------------------------------------------

> summary(FactorConfigFIout)
data:
$`Moderated FI Model.data`
The final iterate satisfies the optimality conditions to the accuracy requested, but the sequence of iterates has not yet converged. NPSOL was terminated because no further improvement could be made in the merit function (Mx status GREEN).

free parameters:
name matrix row col Estimate Std.Error
1 l1 A y1 F1 1.44802605 0.05510917
2 l2 A y2 F1 1.02700153 0.03129520
3 l3 A y3 F1 1.26738868 0.04337198
4 A y1 d_L1 -0.04388849 0.64922665
5 A y2 d_L2 -0.43635869 0.38242596
6 A y3 d_L3 0.14825860 0.55658115
7 A F1 d_F1v 0.09840665 0.40637149
8 A y1 d_r1 -0.05362828 0.04982028
9 A y2 d_r2 -0.44351948 0.06173807
10 A y3 d_r3 0.19566532 0.04162130
11 u1 S y1 y1 2.88522312 0.12568855
12 u2 S y2 y2 0.36705920 0.04068190
13 u3 S y3 y3 1.74437864 0.08607359
14 i1 M 1 y1 6.32458998 0.05755780
15 i2 M 1 y2 7.73262409 0.02995602
16 i3 M 1 y3 6.28814298 0.04359621
17 M 1 d_i1 3.90123137 0.35153257
18 M 1 d_i2 5.48107782 0.48278127
19 M 1 d_i3 3.01220220 0.29538403

observed statistics: 6143
estimated parameters: 19
degrees of freedom: 6124
-2 log likelihood: 22072.42
saturated -2 log likelihood: NA
number of observations: 2354
chi-square: NA
p: NA
AIC (Mx): 9824.423
BIC (Mx): -12736.76
adjusted BIC:
RMSEA: NA
timestamp: 2010-12-02 14:06:20
frontend time: 0.3740001 secs
backend time: 33.819 secs
independent submodels time: 0 secs
wall clock time: 34.193 secs
cpu time: 34.193 secs
openmx version number: 1.0.3-1505

> FactorWeakFI <- mxModel(FactorConfigFI,
+
+ name="FactorWeakFI",
+ mxPath(from=c(latentsY), to=dummies, arrows=1, free=FALSE, values=0),
+ mxData(vmFrame, type="raw")
+ )
>
> FactorWeakFIout <- mxRun(FactorWeakFI)
Running FactorWeakFI
> summary(FactorWeakFI)
observed statistics: 0
estimated parameters: 0
degrees of freedom: 0
-2 log likelihood: NA
saturated -2 log likelihood: NA
number of observations: 0
chi-square: NA
p: NA
AIC (Mx): NA
BIC (Mx): NA
adjusted BIC:
RMSEA: NA
timestamp: NULL
frontend time: NULL
backend time: NULL
independent submodels time: NULL
wall clock time: NULL
cpu time: NULL
openmx version number: NULL

Replied on Fri, 12/03/2010 - 16:16
No user picture. deb193 Joined: 12/01/2010

In reply to by tbrick

but of course! thanks.

the error was in the script I started with, and I just assumed my data was resposible. I should have looked at the output of the script I was given more closely.

The next statement is an omxCompare(), wich had the correct name and the comparison worked. I don;t think anybody ever looked too close at the previous summary, because the comparison was of interest.

... still lots of thanks