In theory you should build models on theory. Most of us are not so lucky to have that.
LASSO is better than stepwise (which is wrong, the use of stepwise that is if all to common). If you want to reduce the number of predictors LASSO or adapted LASSO is preferable - again assuming you have no...
The irony is I have pretty much total permissions in SAS including building permanent tables that are housed on our server (but not the SQl server). That is why I do some long term projects in SAS using Proc SQL. Because I can build permanent tables there and views which I can not on the SQL...
Couldn't you write a Contrast statement to do this? I don't do this type of statistics, and don't use glimnix as a result, but I would look at these statements and see if they help.
Everyone knows that I am learning R finally. :p I was curious what are some good R modules for beginner's. Simple regression is likely to be as complex as I will use. Graphical packages (for someone who will never be a master programmer) or things that interact with SQL would also be nice.
In fact real data that moves in time compared with real data that moves in time can generate extremely high R squared values. That is how you know there is a problem. Time is highly correlated with itself and that is what you are measuring. Not anything substantive.
That is why time series...
Again in my unit the data analyst are all SQL people who do no statistics.
I think meeting someone who has a PHD in statistics is pretty much, except at a University and a handful of government agencies, the same probability as meeting a Siberian Tiger. And that counts doctorates in Psychology...
If two time series, that is two data sets are moving in time and you don't address that you will get absurd R square values that mean nothing. That is the point I was trying to make earlier.
There are no simple solutions for that unfortunately. I don't even pay attention to R square. I don't...
I have read many regression articles over the years (books as well) and the one thing that always puzzles me is the best way to interpret the results? I know slopes and odds ration (and the various measures of the model overall value ) and in honesty they seem pretty limited.
I was wondering...