Hello, I'm currently analyzing antibody data with Repeated Measures ANOVA and have run into problems. I built a model with age, gender, vaccine background, and around 7 genetic polymorphisms. When I do multiple comparisons afterward, I get the error "all pairwise comparisons mean-mean scatterplot cannot be shown because confidence intervals cannot be computed jmp." I don't know how to solve this. Does anybody know what the problem is? Someone suggested something about the degrees of freedom running out, but I do not understand. Appreciate your help!
It's difficult to know without seeing all of your data and your model. Or at least the output for the model from the software. But, running out of degrees of freedom is a likely candidate for the issue. One thing to try is use a model excluding one of your independent variables. It may also be the case that some of your independent variables are co-linear, like if all "male" subjects have the same age.
Hello, one option could be to estimate a model with dichotomous variables in the field of econometrics. This type of model shows results like those of ANOVA. The dichotomous variables will qualify the attributes, in this case, age, gender, etc.
Es posible que exista colinealidad entre las variables, tal vez hay que usar análisis de covarianza. En Regresión usar métodos por pasos. Es complicado estas respuestas sin datos y los corridas de los datos en el software. Saludos
Use Response Surface Methodology (RSM) software to analyse your data. You will definitely have a better result
If you are having trouble running a repeated measures ANOVA in JMP Pro, there could be several reasons for this. Here are some possible explanations and solutions:
Missing Data: Check if there are any missing values in your data. If there are, consider imputing them before running the analysis.
Non-normality: Check if your data is normally distributed. If not, consider transforming the data before running the analysis.
Non-independence: Check if your data violates the assumption of independence. If your data is not independent, you may need to consider alternative methods for analyzing the data, such as a mixed-effects model.
Incorrect Input: Make sure you are selecting the correct variables and options for your analysis.
I suggest you to use either the SAS PROC MIXED procedure, whic allow you to moxel the covariance stru ture for the repeated measures factor.
Perhaps your problem is with the estimate of a variance component without the printout is very difficult to say more.
If you can't use sas look on the software wether you can model the covariance structure....
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs