Skip to first unread message
Anusha Mishra
Jun 11, 2019, 5:05:08 PM6/11/19
to lavaan
I'm a complete novice to LAVAAN and advanced statistics in general.
My SEM Analysis is giving the following errors:
My SEM Analysis is giving the following errors:
Warning messages:
1: In lav_data_full(data = data, group = group, cluster = cluster, :
lavaan WARNING: some observed variances are (at least) a factor 1000 times larger than others; use varTable(fit) to investigate
2: In lav_model_vcov(lavmodel = lavmodel, lavsamplestats = lavsamplestats, :
lavaan WARNING:
Could not compute standard errors! The information matrix could
not be inverted. This may be a symptom that the model is not
identified.
3: In lav_object_post_check(object) :
lavaan WARNING: some estimated ov variances are negative
4: In lavaan::lavaan(model = final.model, data = tryfour, model.type = "sem", :
lavaan WARNING: not all elements of the gradient are (near) zero;
the optimizer may not have found a local solution;
use lavInspect(fit, "optim.gradient") to investigate
Additional Information:
Additional Information:
lavaan 0.6-3 ended normally after 4513 iterations
Optimization method NLMINB
Number of free parameters 87
Number of observations 85920
Estimator ML
Model Fit Test Statistic 264379.635
Degrees of freedom 255
P-value (Chi-square) 0.000
Model test baseline model:
Minimum Function Test Statistic 471891.712
Degrees of freedom 330
P-value 0.000
User model versus baseline model:
Comparative Fit Index (CFI) 0.440
Tucker-Lewis Index (TLI) 0.275
Loglikelihood and Information Criteria:
Loglikelihood user model (H0) -607010.903
Loglikelihood unrestricted model (H1) -474821.085
Number of free parameters 87
Akaike (AIC) 1214195.805
Bayesian (BIC) 1215010.227
Sample-size adjusted Bayesian (BIC) 1214733.739
Root Mean Square Error of Approximation:
RMSEA 0.110
90 Percent Confidence Interval 0.109 0.110
P-value RMSEA <= 0.05 0.000
Standardized Root Mean Square Residual:
SRMR 0.071
Parameter Estimates:
Information Expected
Information saturated (h1) model Structured
Standard Errors Standard
> varTable(final.fit)
name idx nobs type exo user mean var nlev lnam
1 web2 43 85920 numeric 0 0 0.899 0.091 0
2 web3 44 85920 numeric 0 0 0.038 0.037 0
3 web4 45 85920 numeric 0 0 0.006 0.006 0
4 web5 46 85920 numeric 0 0 0.045 0.043 0
5 DELIVER 47 85920 numeric 0 0 3.095 21.665 0
6 phone2 18 85920 numeric 0 0 0.736 0.194 0
7 phone3 19 85920 numeric 0 0 0.052 0.050 0
8 phone4 20 85920 numeric 0 0 0.011 0.011 0
9 phone5 21 85920 numeric 0 0 0.178 0.146 0
10 CarsPerDri 8 85920 numeric 0 0 1.125 0.345 0
11 PTUSE 48 85920 numeric 0 0 1.199 25.921 0
12RIDESHARE 39 85920 numeric 0 0 0.702 9.678 0
13 msa_pt 51 85920 numeric 1 0 0.039 0.004 0
14 HHSIZE 40 85920 numeric 1 0 2.163 1.407 0
15 income2 11 85920 numeric 1 0 0.044 0.042 0
16 income3 12 85920 numeric 1 0 0.126 0.110 0
17 income4 13 85920 numeric 1 0 0.317 0.216 0
18 income5 14 85920 numeric 1 0 0.234 0.179 0
19 income6 15 85920 numeric 1 0 0.069 0.065 0
20 homeown2 5 85920 numeric 1 0 0.254 0.189 0
21 homeown3 6 85920 numeric 1 0 0.007 0.006 0
22 educ2 36 85920 numeric 1 0 0.022 0.021 0
23 educ3 37 85920 numeric 1 0 0.134 0.116 0
24 educ4 38 85920 numeric 1 0 0.556 0.247 0
25 Age2 30 85920 numeric 1 0 0.022 0.021 0
26 Age3 31 85920 numeric 1 0 0.121 0.107 0
27 Age4 32 85920 numeric 1 0 0.235 0.180 0
28 Age5 33 85920 numeric 1 0 0.317 0.216 0
29 Race2 25 85920 numeric 1 0 0.082 0.076 0
30 Race3 26 85920 numeric 1 0 0.043 0.042 0
31 Race4 27 85920 numeric 1 0 0.111 0.099 0
32 CIMean:YearsMean 54 85920 numeric 1 0 6.368 2538.760 0
33 YearsMean 53 85920 numeric 1 0 -0.041 3.726 0
34 CIMean 50 85920 numeric 1 0 0.000 557.780 0
Can anyone please help?
nick judd
Jun 11, 2019, 5:35:27 PM6/11/19
to lav...@googlegroups.com
First deal with the variances I've realized sometimes that solves everything else. Do as it says vartable(fit) and multiple or divide the offending variable by a constant (e.g. 10 or 100).
Essentially it makes it hard for the optimizer to find a solution. There are previous posts that explain it better.
Best,
Nick
--
You received this message because you are subscribed to the Google Groups "lavaan" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lavaan+un...@googlegroups.com.
To post to this group, send email to lav...@googlegroups.com.
Visit this group at https://groups.google.com/group/lavaan.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/41dc4647-f96c-4bd4-9ff2-11cec3f514c4%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Reply all
Reply to author
Forward
0 new messages