Recent content by hlsmith

Your context is different from mine. Mine will usually be medical patients in the dataset more than once (so obs clustered in patients) or patients clustered in hospitals. Controlling for it, allows you to control for the between and within group variability and define if there is a difference...

Wait - does this mean people could contribute more than one observation, and that the values would be correlated. If so, fitting an empty model controlling for the covariance structure is necessary. I usually liken this to allowing a person to vote twice or more. In the non-descriptive - but...

I guess it depends on the purpose and how you are using the results. I believe in the past, I have fit empty models but controlled for the clusters. Can you tell us more about your context.

Yes that is the package. Welcome to the forum!

You will likely have issues. I would conduct analyses using your own data and make superficial comparisons in your discussion section. Even if you had the data, there could be missing confounders or selection bias making the sets not comparable. Of note, I know in R software there is a package...

Post the output from the model and we can help. But all coefficients in in reference to your base case (e.g., intercept). Some programs will let you define effect estimates to get at the other terms. SAS: LSMEANS R: emeans

I have never seen a correction for having an interaction and its main terms in the model.

If the IV is deterministically created from the other variables, they wouldn't contribute anything new. But you should tell us more, if you want additional insight.

So where are these data coming from. How are you simulating them and the source of the reference? Are they supposed to be counts?

That is how I read it. It it makes sense - but I hadn't thought about it that way.

Yes, you could just use regression for a ttest here and examine residuals. Have you run many regressions? Because if you have, you will note that the residuals can help discern a non-linear relationship, wrong defined data generating function, or heterogeneity resulting in a funnel shape in...

I hadn't heard this before. Can you elaborate.

P.S., This concept would fall under bias/variance trade off. Less variables means more potential bias and more variables (given finite sample) means greater standard errors. If you have a loss function or value you are trying to optimize (e.g., MSE or accuracy) -- having a random holdout set...

You need to create a table with the listed variables in the model along with the AIC values and share it - selection based on pvalues isn't always great. Model reduction can also be investigated using Least absolute shrinkage selection operator based regression. Though it is important to...

@Miner