Modeling a quadratic term with three ... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Tim Seifert posted on Wednesday, February 21, 2007 - 7:09 am
Hello,
The data I would like to model is children's recognition of printed words. At pretest (t=0), posttest (2 months), and follow-up (8 months) the number of words children could identify was recorded. The pretest data was normally distributed, a little less so at the other time points. Two questions.

1) Should the data be modelled as Poisson or continuous distributions? If modelled as a continuous (normal) distribution, the intercept = the meannumber of words at pretest (5.5). When modelled as Poisson, I am not sure what the intercept represents since its value is about 1.3.

2) Examination of the data suggests a quadratic term would be useful. I noticed in previous posts that four time points are preferred to three, and in adding a quadratic term the number of free parameters is zero. But I tried constraining the intercept and linear means, variances and covariances, but still have no free parameters. Can a quadratic term be modelled with only three time points?
 Bengt O. Muthen posted on Saturday, February 24, 2007 - 4:17 pm
1) With high counts, Poisson is approximately normal so I would go with normal since it is simpler to work with. The Poisson means are on a log rate scale where rate is the "lambda" parameter of a Poisson distribution (see stat texts),

2)You can have a quadratic even with 3 time points, but you can't have all of its potential var-cov parameters free. You can for example have a random intercept and random linear slope but a fixed quadratic. That is just-identified. There is no use in restraining the model just to get df's.
 Johannes Meier posted on Sunday, July 31, 2011 - 11:13 am
Dear all,

I have estimated a "Fixed Quadratic, Random Linear Model" in Mplus using the following multilevel syntax:

! time-level model;

%WITHIN%
linear | y ON t_linear;
quadratic | y ON t_quadratic;

! person-level model;

%BETWEEN%
y*
linear*;
y WITH linear;
quadratic@0;

However, I didn't figure out how to specify this model using the Mplus growth curve syntax (i.e. i s q | y1@0 y2@1 y@2) . I would greatly appreciate your help.

Thanks,
Johannes Meier
 Linda K. Muthen posted on Monday, August 01, 2011 - 10:50 am
i s q | y1@0 y2@1 y@2

is the way to specify a quadratic growth model. The problem is that it is not identified with three time points. You need at least four.
 Johannes Meier posted on Monday, August 01, 2011 - 2:39 pm
Hi Linda,

Thanks for your answer.

I have over 20 time points and I just wondered how to specify a "Fixed Quadratic, Random Linear Model" in the regular MPlus growth curve syntax. In my example above I did it the multilevel way (which I am used to as a former Stata user).

If I would use just
i s q | y1@0 y2@1 y3@2 ... y20@19
I would estimate a "Random Quadratic, Random Linear Model".

Do you have any hint for me?

Best regards,
Johannes
 Linda K. Muthen posted on Tuesday, August 02, 2011 - 8:06 am
i s q | y1@0 y2@1 y3@2 ... y20@19
q@0;

I would divide the time scores by ten given that the quadratic times scores are the squared value of the linear time scores.
 EFried posted on Monday, February 06, 2012 - 8:04 am
Dear Dr Muthén,

when adding a quadratic term

i s q | y1@0 y2@1 y3@2

Does the term "q" also have to be added to the

i s (Q) ON x1 x2 x3;

and the

c%1
y1 y2 y3 i s (Q?);

statements?

Thank you
 Linda K. Muthen posted on Monday, February 06, 2012 - 9:11 am
Yes, in theory, but a quadratic growth model is not identified with three time points.
 EFried posted on Wednesday, February 08, 2012 - 6:20 am
I have 5 time points and just copied the example from above - apologies for the confusion.
Thank you for the answer, I'll include the quadratic term in my ".. ON .." statements!
 sed48 posted on Monday, June 12, 2017 - 4:56 am
Dear Muthen,

I have 6 timepoints in my dataset. I have tried a variety of models, including linear, quadratic, and piecewise models (with 1 breaking point). I have tried two types of piecewise models, either where both slopes are linear, or where one is quadratic and the other linear. My question concerns the piecewise model with both a linear segment and quadratic segment.

i s1 q1 | y0@0 y1@1 y2@2 y3@3 y4@3 y5@3

i s2 | y0@0 y1@0 y2@0 y3@0 y4@4 y5@5

I am wondering whether this is really too complicated a model for the number of timepoints I have and it would result is overfitting.

I would appreciate your advice on this.

Many thanks!
 Bengt O. Muthen posted on Monday, June 12, 2017 - 6:10 pm
I would think that piecewise is more motivated when there is a substantive reason for a change in the growth. Perhaps you can instead try a cubic.
 Marco Marinucci posted on Thursday, June 18, 2020 - 12:02 am
Dear all,
would it make sense to constrain the residual variance of the observed variable to zero to allow the estimation of the quadratic term with three timepoints?

I am more interested in the actual trajectory of the data, and an initial observation suggests that the quadratic curve is best representing the data. Would alternatively make sense if I would estimate a quadratic term without a linear one?
Thank you!
 Bengt O. Muthen posted on Thursday, June 18, 2020 - 4:09 pm
Q1: No.

Q2: No.

Equal residual variances over time and no quadratic variance might do it.
 Marco Marinucci posted on Thursday, June 18, 2020 - 11:23 pm
Dear Dr. Muthen,
thank you for your prompt answer.
My research questions look at predictors of the slopes (linear and quadratic). I think I may not be able to answer them if I set no variation in the quadratic slope.

Do you have any hint on the best statistical strategy to adopt?

Thank you in advance!
 Bengt O. Muthen posted on Friday, June 19, 2020 - 6:16 pm
You can regress quadratic on covariates even if the quadratic residual variance is zero. But a quadratic for 3 time points is always going to be difficult to defend.

You may want to discuss this on SEMNET.
 ksk posted on Tuesday, June 30, 2020 - 1:04 pm
Dear Dr. Muthen,

I'm trying to fit a quadratic growth model using 6 time points. I want to set the intercept at the third time point. I used the command below:

i s q| GPK@-2 GK@-1 G1@0 G2@1 G3@2 G5@4;

However, the loadings on the quadratic slope didn't look correct. It was shown 4, 1, 0, 1, 4, and 16. So, I generated a new command below and it seems correct:

i s| GPK@-2 GK@-1 G1@0 G2@1 G3@2 G5@4;
i q| GPK@-4 GK@-1 G1@0 G2@1 G3@4 G5@16;

Would you please let me know which command should I use?

Thank you!
 Bengt O. Muthen posted on Wednesday, July 01, 2020 - 4:13 pm
Your statemment

i s q| GPK@-2 GK@-1 G1@0 G2@1 G3@2 G5@4;

says that the "s" loadings have the values you give. The "q" values are the square of those. Keep this way of doing it.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: