I’ve run a lot of linear regressions. The first thing that jumps out at me are the initial studies conducted, which pull the tail fit up quite a lot.
there are a huge number of methods to test for stability of the model. Does anyone see a way to access the model data?
Beyond the model, knowing the distribution of sperm counts across men and across men of different races would be immensely informative. increasing sample sizes or simple increasing total observations could actually make it look like a change is occurring when in reality you are simply converging to the global population mean.
there are a huge number of methods to test for stability of the model. Does anyone see a way to access the model data?
Beyond the model, knowing the distribution of sperm counts across men and across men of different races would be immensely informative. increasing sample sizes or simple increasing total observations could actually make it look like a change is occurring when in reality you are simply converging to the global population mean.