Kentuckian for Joe Biden kentucky democrats shirt
As you have classified this under regression statistics, I presume you have related variables. The first thing is to decide which variables are explanatory variables and which is the response. The next is to plot the data. With only one explanatory variable use one of SASs plotting procs or one from R (which many people think is superior to those in SAS). With more than one explanatory variable you need a matrix plot (I think that's what it's called) that plots all different pairs of variables.
Once you get a feel for the data and whether regressions are linear, you can start doing your regression analysis. Look out for outliers and influential values that can give misleading results. One nice thing about some of the more recent SAS procs is that they can be made to give a range of diagnostics automatically.
Once you get a feel for the data and whether regressions are linear, you can start doing your regression analysis. Look out for outliers and influential values that can give misleading results. One nice thing about some of the more recent SAS procs is that they can be made to give a range of diagnostics automatically.
Nhận xét
Đăng nhận xét