Interpreting GBTA outout

[Hayoung Choi]

Hello,

If someone could help me with interpreting the output of group based trajectory analysis in Stata
The outout reads as follows:

. traj, var(stability*) indep(year*) model(cnorm) order(3 0 0 0) min(0) max(4) //

==== traj stata plugin ==== Jones BL Nagin DS, build: Mar 17 2021

901 observations read.
901 observations used in the trajectory model.

Maximum Likelihood Estimates
Model: Censored Normal (cnorm)

Standard T for H0:
Group Parameter Estimate Error Parameter=0 Prob > T

1 Intercept 1.87518 6682.47446 0.000 0.9998

Linear -9.96981 10.61092 -0.940 0.3475

Quadratic 0.00986 0.00516 1.912 0.0560

Cubic -0.00000 0.00000 -3.196 0.0014

2 Intercept 3.58373 0.17708 20.238 0.0000

3 Intercept 5.06939 0.12004 42.231 0.0000

4 Intercept 13.20601 44.31632 0.298 0.7657

Sigma 1.66693 0.02857 58.350 0.0000

Group membership
1 (%) 27.26919 2.03163 13.422 0.0000
2 (%) 15.38789 2.26282 6.800 0.0000
3 (%) 21.99592 2.47955 8.871 0.0000
4 (%) 35.34700 1.71763 20.579 0.0000

BIC= -6702.83 (N=7311) BIC= -6691.32 (N=901) AIC= -6664.90 ll= -6653.90



As you can note in the syntex, I designated the first group to have a parameter of Cubic, and the result was only significant (p=0.0014) in cubic and then quadratic.
Is it OK that the significance is lower than 0.1 or 0.05 only in Cubic when intercept, and linear were not in the first group? since I designated the group to be Cubic?

Moreover, if only the fourth group were statistically significant (when now is not) this trajectory group will be statistically significant overall?

[Hayoung Choi]

ohhh it was GBTM