Degrees of freedom, according to my understanding should be n-1. But as I am observing in most of the questions (Schweser question bank), they have taken them as n-2. Any comment or clarification, would help me.

The number of degrees of freedom depends on how many parameters you’ve estimated.

In a multiple regression, for example, in which you have *n* data points and *k* independent variables and you’re estimating *k* slope coefficients and one intercept, *dof* = *n* − *k* − 1.

What are types of problems in which you’re seeing *dof* = *n* − 2?

These are the kinds of things I can’t believe the prep providers and the CFAI fail to explain or even give the first sentence that s2000 provided. From the whacky misconception that multicollinearity inflates R^{2 }or the F-statistic to the “mysteriously new” t-statistic for Pearson’s correlation coefficient… they look like a deer on ice.