Quiz — Excel Functions for Regression 15 T/F

1) =SLOPE(known_ys, known_xs) returns the estimated slope \( \beta_1 \).

2) =INTERCEPT(known_ys, known_xs) returns the estimated intercept \( \beta_0 \).

3) =RSQ(x_range, y_range) returns the correlation \( r \).

4) =CORREL(x_range, y_range) equals the slope.

5) =STEYX(known_ys, known_xs) approximates \( \sqrt{\text{MSE}} \), the typical vertical error in Y-units.

6) =FORECAST.LINEAR(x0, known_ys, known_xs) computes the predicted \( \hat y \) at \( x_0 \).

7) If \( \beta_0 \) is in E2 and \( \beta_1 \) in E3, a residual in row 2 is =B2 - ($E$2 + $E$3*A2).

8) The slope from LINEST (with TRUE,TRUE) can be pulled by =INDEX(LINEST(Y,X,TRUE,TRUE),1,1).

9) The degrees of freedom for inference are returned by =INDEX(LINEST(Y,X,TRUE,TRUE),4,1).

10) A two-sided p-value for slope can be computed as =T.DIST.2T(ABS(t), df), where t = slope / SE(slope).

11) Residuals should be absolute differences: =ABS(actual - predicted).

12) In simple regression, \( R^2 \) can be computed via =CORREL(Y,X).

13) Using only functions, you cannot derive ANOVA pieces (SSE, SSR, MSE) without ToolPak.

14) When pasting tab-separated data into Excel, “Text to Columns → Delimited → Tab” can fix a single-column paste.

15) The argument order in =SLOPE doesn’t matter; =SLOPE(X,Y) equals =SLOPE(Y,X).