Azzera filtri
Azzera filtri

Confidence intervals for fitlme

19 visualizzazioni (ultimi 30 giorni)
jjohn33
jjohn33 il 2 Feb 2020
Risposto: Myrtle42 il 5 Ago 2020
How can I plot the confidence intervals from the fitlme mixed model?
lme = fitlme(T1,' par ~ 1 + cov + (1 I ID)','FitMethod','REML');
Example:
Image 8.jpg
Results:
Fixed effects coefficients (95% CIs):
Name Estimate SE tStat DF pValue Lower Upper
{'(Intercept)'} 0.021452 0.0022428 9.5647 50 7.0315e-13 0.016947 0.025957
{'cov' } 0.00022556 0.00017193 1.3119 50 0.19554 -0.00011977 0.0005709

Risposta accettata

jjohn33
jjohn33 il 3 Feb 2020
The plot image is only showing for standard regression not included random for the intercept in this case. The lme. Coefficient is only one value and it is not calculated over the prediction line.

Più risposte (2)

Jakob B. Nielsen
Jakob B. Nielsen il 3 Feb 2020
If you look at the linear mixed model documentation you should be able to extract the confidence interval values from lme.Coefficients.Upper and lme.Coefficients.Lower. (They are the red dotted lines already in your plot, by the way).

Myrtle42
Myrtle42 il 5 Ago 2020
In case anyone else is still wondering about this, the predict function is the way to go:

Prodotti


Release

R2019b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by