Most analysts’ primary focus is to check the distributional assumptions with regards to residuals. They must be independent and identically distributed (i.i.d.)
with a mean of zero and constant variance.
Residuals can also give us insight into the quality of our models.
In this webinar, we’ll review and compare what residuals are in linear regression, ANOVA, and generalized linear models. Jeff will cover:
Knowing how to piece this information together will improve your statistical modeling skills.
Note: This training is an exclusive benefit to members of the Statistically Speaking Membership Program and part of the Stat’s Amore Trainings Series. Each Stat’s Amore Training is approximately 90 minutes long.
One question that seems to come up pretty often is:
Well, let’s start with how they’re the same:
Both are types of generalized linear models. This means they have this form:
Proportion and percentage data are tricky to analyze.
Much like count data, they look like they should work in a linear model.
They’re numerical. They’re often continuous.
And sometimes they do work. Some proportion data do look normally distributed so estimates and p-values are reasonable.
But more often they don’t. So estimates and p-values are a mess. Luckily, there are other options. (more…)
I received a question recently about R Commander, a free R package.
R Commander overlays a menu-based interface to R, so just like SPSS or JMP, you can run analyses using menus. Nice, huh?
The question was whether R Commander does everything R does, or just a small subset.
Unfortunately, R Commander can’t do everything R does. Not even close.
But it does a lot. More than just the basics.
So I thought I would show you some of the things R Commander can do entirely through menus–no programming required, just so you can see just how unbelievably useful it is.
Since R commander is a free R package, it can be installed easily through R! Just type install.packages("Rcmdr") in the command line the first time you use it, then type library("Rcmdr") each time you want to launch the menus.
Import data sets from other software:
Define Numerical Variables as categorical and label the values
Open the data sets that come with R packages
Merge Data Sets
Edit and show the data in a data spreadsheet
Personally, I think that if this was all R Commander did, it would be incredibly useful. These are the types of things I just cannot remember all the commands for, since I just don’t use R often enough.
Yes, R Commander does many of the simple statistical tests you’d expect:
What is surprising though, is how many higher-level statistics and models it runs:
In other words–you can use R Commander to run in R most of the analyses that most researchers need.
A sample of the types of graphs R Commander creates in R without you having to write any code:
The nice part is that it does not only do simple versions of these plots. You can, for example, add regression lines to a scatter plot or run histograms by a grouping factor.
If you’re ready to get started practicing, click here to learn about making scatterplots in R commander, or click here to learn how to use R commander to sample from a uniform distribution.
In my last blog post we fitted a generalized linear model to count data using a Poisson error structure.
We found, however, that there was over-dispersion in the data – the variance was larger than the mean in our dependent variable.