AP Statistics · Topic 2.7

Residuals Practice

Part of Exploring Two-Variable Data.(DAT-1.D)

Practice questions

9

Want a predicted score for the whole AP STAT exam? Take the 20-question diagnostic and Lumi will plan the rest.

Sample questions

5 of 9 — sign in to practice the rest with adaptive difficulty and mastery tracking.

  1. Sample 1difficulty 1/5

    A regression model predicts y-hat = 8 for an observation whose actual y-value is 11.

    x (units) y (units) obs y=11 y-hat=8

    What is the residual?

    • A

      3

      check_circle
    • B

      11

    • C

      -3

    • D

      8

    Why

    Residual = observed - predicted = 11 - 8 = 3.

  2. Sample 2difficulty 2/5

    A regression predicts y-hat = 2x + 5. For x = 6, the observed y is 14.

    Regression line and observation y-hat = 2x + 5 Observation: x = 6, y = 14 Residual = ?

    What is the residual?

    • A

      -17

    • B

      -3

      check_circle
    • C

      17

    • D

      3

    Why

    y-hat = 2(6) + 5 = 17. Residual = 14 - 17 = -3.

  3. Sample 3difficulty 2/5

    For the line y-hat = 10 + 3x with observation (x=4, y=20).

    y-hat = 10 + 3x; observation (x=4, y=20) Predicted: 10 + 3(4) = 22 Observed: 20

    What is the residual?

    • A

      20

    • B

      22

    • C

      -2

      check_circle
    • D

      2

    Why

    Residual = observed - predicted = 20 - 22 = -2.

  4. Sample 4difficulty 2/5

    Consider the residuals from a least-squares regression line.

    Property of LSRL residuals Sum of residuals = ? (Least-squares with intercept)

    Which statement is always true?

    • A

      All residuals are positive

    • B

      Residuals equal predicted values

    • C

      The sum of squared residuals equals 0

    • D

      The sum of the residuals equals 0

      check_circle

    Why

    The least-squares method produces residuals that sum to 0 (when an intercept is included).

  5. Sample 5difficulty 3/5

    Output for predicting weight from height shows residual SD s = 10.

    Predicting weight from height y-hat = -120 + 4.5 x s = 10 lb (residual SD) R-Sq = 64%

    Which interpretation of s is best?

    • A

      The line passes within 10 inches of every point

    • B

      Typical prediction error is about 10 lb

      check_circle
    • C

      10 lb equals the slope of the line

    • D

      About 10% of observations are below the line

    Why

    s estimates the typical (root-mean-square) size of residuals, i.e., the typical prediction error.