Statistics · Regression · Hard
Residuals against y
You regress y on x with an intercept and get an R² of 0.64. You then regress the residuals of that fit on y, again with an intercept. What is the slope of this second regression?
a fraction, an expression or a decimal to two significant figures · 5/8 · ln 2 · pi · inf
Answer and solution
Answer
9/25 ≈ 0.36
Solution
Residuals are uncorrelated with the fit but not with y itself: their covariance with y is their own variance.
cov(e, y) = cov(e, ŷ + e) = var(e) = (1 − R²) var(y). Slope = (1 − 0.64) = 0.36.
Regression comes up in 11 firms' roundsHudson River Trading · Goldman Sachs · D. E. Shaw · +8
These 16 rounds ask this kind of problem live, with follow-ups. Every round on every track is in Premium.