“If there is a strong relationship between the variables and the SSE is small, the individual estimation errors will also be small.” Can someone explain to me what effect the relationship between the variables play in that statement?
SSE is the regression residual errors summed up for you. If the fit is good, then SSE makes a small contribution to the total variation of the variable, and the rest is explained by the RSS, which is the line you created (this is the definition of coefficient of variation). Less SSE --> more relative RSS, more predictable is the dependent variable from your model. Badaboom, badabing, badabang Think of SSE --> this is variation around your predicted line (your residuals/errors)
when they say a strong relationship between the variables, do they mean the relationship of all the X to Y, causing a high RSS?
a strong relationship for x to y will indicate high explanatory power (r^2). A high explanatory power with a low SSE, then you have a pretty solid model. Errors will be small.
alright, thanks. i just wasnt sure of what they meant by the relationship between the variables. i thought it could be either X to Y or multicollinearity.