Skip to main content

New answers tagged

1 vote
Accepted

Multilevel Model Residuals Scatterplot Assumptions

You only have a fixed number of outcomes, so the pattern you are observing reflects that. In fact, you can quite closely reproduce this pattern by plotting residuals of a binomial process: ...or the ...
Frans Rodenburg's user avatar
0 votes

Creating ROC curve for multi-level logistic regression model in R

This question is super old, but for those coming to it now I believe the author is purposely referring to multi-level models, not multi-class models. Some info regarding multi-level models: https://...
Kelley Brady's user avatar
0 votes

Mixed Model for Repeated Measurement (mmrm) - Assumptions

thanks for your question, I just found it by chance - I would recommend considering posting questions specifically on mmrm in the corresponding GitHub issues page ...
Daniel Sabanes Bove's user avatar
6 votes
Accepted

What is the mean of random effects?

In mixed model terminology, a random effect is typically a normally distributed effect with mean zero and some unknown variance to be estimated: $$ \begin{aligned} \mathbf{y} &= \mathbf{X}\mathbf{\...
Frans Rodenburg's user avatar
1 vote
Accepted

Centering Variables in Multilevel Models with Longitudinal Data

The Enders & Tofighi (2007) article is fantastic. I do think that they are light on details for the longitudinal case because that wasn't the focus of their paper. Fortunately, there are other ...
Erik Ruzek's user avatar
  • 5,390
1 vote

Lack of within-cluster variability

Assuming you are using a mixed model, i.e. model clusters with random effects, even if every hospital uses only one treatment it will be fine, as long as you have enough hospitals. (See here: https://...
Lukas Lohse's user avatar
  • 2,842
3 votes
Accepted

Longitudinal analysis of peer effect

Your proposed multilevel model for examining the effect of the intervention on sibling influence on outcomes looks OK to me. We can break down and critically evaluate each part: Model Specification: \...
Robert Long's user avatar
  • 63.9k
1 vote
Accepted

How can I regress group level outcomes on individual characteristics?

As @Stefan alluded to, your best bet is to utilize the multilevel structural equation modeling paradigm. Software for this can be found both in existing general purpose programs/packages (...
Erik Ruzek's user avatar
  • 5,390
1 vote

Mixed models - Intuition of correlated discrete random effects

They provide the variance covariance plot which seems to show the covariance between each pair of day and subject. Positive correlation implies that by increasing the var1 (Days) the var2 increasing ...
Robert Long's user avatar
  • 63.9k
3 votes

Mixed models - Intuition of correlated discrete random effects

If we name the variables in your model as follows reaction = a + b * days + (c + d * Days) * Subject, the correlation is the correlation between ...
Gijs's user avatar
  • 3,889
4 votes

What is the difference between a) multilevel modelling and b) adding a categorical IV to a multiple regression?

Let's say you have a dataset of children grouped into classes. One way to analyze these data is to treat "class" as just another (categorical) independent variable - as a characteristic of ...
Graham Wright's user avatar
3 votes

What is the difference between a) multilevel modelling and b) adding a categorical IV to a multiple regression?

In the situation you describe, with a single grouping variable, practically speaking it can indeed be effectively the same when the grouping variable enters the model as a fixed effect (independent ...
Robert Long's user avatar
  • 63.9k
2 votes
Accepted

Why do top-down approaches produce biased coherent forecasts?

Think about the simplest possible non-trivial hierarchy, with one aggregate series $y_t$, and two bottom-level disaggrete series $a_t$ and $b_t$, where $y_t = a_t+b_t$. Then $$ S = \begin{bmatrix}1 &...
Rob Hyndman's user avatar
  • 57.5k
2 votes

Three-tier Multi-Level Model: violation of the assumptions of normality and heteroscedasticity

With values ranging from zero to 100, such a response variable can often be considered continuous. You mentioned a gamma model, but the support for the gamma distribution is $(0, \infty)$. A good ...
Robert Long's user avatar
  • 63.9k

Top 50 recent answers are included