15 extra multiple-choice questions for Regression Analysis (12th Standard Statistics, Samacheer Kalvi), beyond the ones printed in the textbook — each with the correct option highlighted and a clear, worked explanation. Free to read in English and Tamil.
Q1
If \( b_{YX} = 0.8 \) and \( b_{XY} = 0.45 \), the correlation coefficient \( r_{XY} \) is
- A. 0.36
- B. 0.625
- C. 0.6Correct
- D. \( -0.6 \)
Explanation. \( r_{XY} = \sqrt{b_{XY} \times b_{YX}} = \sqrt{0.45 \times 0.8} = \sqrt{0.36} = 0.6 \). It is positive because both regression coefficients are positive. 0.36 is \( r^2 \) and 0.625 is their arithmetic mean.
Q2
Given \( r_{XY} = 0.5 \), \( SD(X) = 4 \) and \( SD(Y) = 10 \), the regression coefficient of \( Y \) on \( X \) is
- A. 0.2
- B. 1.25Correct
- C. 2.5
- D. 20
Explanation. \( b_{YX} = r_{XY} \frac{SD(Y)}{SD(X)} = 0.5 \times \frac{10}{4} = 1.25 \). The value 0.2 is \( b_{XY} = 0.5 \times \frac{4}{10} \), and 2.5 forgets to multiply by \( r_{XY} \).
Q3
The two regression lines of a bivariate data are \( x + 2y = 10 \) and \( 2x + y = 11 \). The mean values \( (\bar{x}, \bar{y}) \) are
- A. (3, 4)
- B. (10, 11)
- C. (5, 5.5)
- D. (4, 3)Correct
Explanation. Both regression lines pass through \( (\bar{x}, \bar{y}) \), so solve them together. Doubling the first: \( 2x + 4y = 20 \); subtracting the second gives \( 3y = 9 \), so \( \bar{y} = 3 \) and \( \bar{x} = 10 - 6 = 4 \).
Q4
For a bivariate data, \( \bar{x} = 20 \), \( \bar{y} = 30 \) and \( b_{YX} = 1.5 \). Using the regression equation of \( Y \) on \( X \), the estimate of \( Y \) when \( x = 24 \) is
- A. 36Correct
- B. 66
- C. 24
- D. 33
Explanation. Slope-point form: \( \hat{Y} - \bar{y} = b_{YX}(x - \bar{x}) \). So \( \hat{Y} = 30 + 1.5(24 - 20) = 30 + 6 = 36 \). The value 66 wrongly uses \( 1.5 \times 24 \) without subtracting \( \bar{x} \).
Q5
The mean and SD of the yield \( Y \) are 50 and 6; the mean and SD of rainfall \( X \) are 20 and 3; \( r_{XY} = 0.6 \). The estimated yield when rainfall is 25 is
- A. 51.5
- B. 56Correct
- C. 60
- D. 53
Explanation. \( b_{YX} = r_{XY} \frac{SD(Y)}{SD(X)} = 0.6 \times \frac{6}{3} = 1.2 \). Then \( \hat{Y} = 50 + 1.2(25 - 20) = 50 + 6 = 56 \). Using \( b_{XY} = 0.3 \) gives 51.5; dropping \( r_{XY} \) gives 60.
Q6
For \( n = 5 \) pairs, \( \sum x_i = 15 \), \( \sum y_i = 40 \), \( \sum x_i^2 = 55 \) and \( \sum x_i y_i = 140 \). Solving the normal equations, the fitted regression equation of \( Y \) on \( X \) is
- A. \( \hat{Y} = 8 + 2x \)
- B. \( \hat{Y} = 2 + 0.5x \)
- C. \( \hat{Y} = 4 + 2x \)
- D. \( \hat{Y} = 2 + 2x \)Correct
Explanation. Normal equations: \( 5a + 15b = 40 \) and \( 15a + 55b = 140 \). Three times the first is \( 15a + 45b = 120 \); subtracting gives \( 10b = 20 \), so \( b = 2 \) and \( a = (40 - 30)/5 = 2 \). Hence \( \hat{Y} = 2 + 2x \).
Q7
For \( n = 4 \), \( \sum x = 10 \), \( \sum y = 20 \), \( \sum xy = 58 \) and \( \sum x^2 = 30 \). Using \( b_{YX} = \frac{n\sum xy - \sum x \sum y}{n\sum x^2 - (\sum x)^2} \), the value of \( b_{YX} \) is
- A. 0.8
- B. 1.93
- C. 1.6Correct
- D. 3.2
Explanation. Numerator: \( 4 \times 58 - 10 \times 20 = 232 - 200 = 32 \). Denominator: \( 4 \times 30 - 10^2 = 120 - 100 = 20 \). So \( b_{YX} = \frac{32}{20} = 1.6 \). The value 1.93 is simply \( \frac{58}{30} \).
Q8
The two regression equations are \( \hat{Y} = 20 - 0.4x \) and \( \hat{X} = 30 - 0.9y \). The correlation coefficient between \( X \) and \( Y \) is
- A. \( -0.6 \)Correct
- B. 0.6
- C. \( -0.36 \)
- D. \( -0.65 \)
Explanation. \( b_{YX} = -0.4 \) and \( b_{XY} = -0.9 \). Then \( r^2 = (-0.4)(-0.9) = 0.36 \), so \( r = \pm 0.6 \). Both coefficients are negative, so \( r_{XY} = -0.6 \). The value \( -0.65 \) is their arithmetic mean.
Q9
Given \( V(X) = 25 \), \( b_{XY} = 0.5 \) and \( r_{XY} = 0.6 \), the standard deviation of \( Y \) is
- A. 30
- B. 6Correct
- C. 4.17
- D. 1.5
Explanation. \( SD(X) = \sqrt{25} = 5 \). From \( b_{XY} = r_{XY} \frac{SD(X)}{SD(Y)} \): \( 0.5 = 0.6 \times \frac{5}{SD(Y)} \), so \( SD(Y) = \frac{3}{0.5} = 6 \). Using the variance 25 instead of the SD gives 30.
Q10
Which of the following pairs can be the regression coefficients \( (b_{YX}, b_{XY}) \) of the same data?
- A. (1.6, 0.9)
- B. \( (-0.5, 0.8) \)
- C. (0.8, 0.45)Correct
- D. (2, 0.6)
Explanation. Both coefficients must have the same sign and their product \( r^2 \) cannot exceed 1. (0.8, 0.45) gives \( r^2 = 0.36 \). (1.6, 0.9) and (2, 0.6) give products 1.44 and 1.2; \( (-0.5, 0.8) \) has opposite signs.
Q11
The method of least squares finds the line of best fit by making
- A. the sum of the residuals the largest
- B. the sum of the observed values zero
- C. the sum of the squares of the observed values the least
- D. the sum of the squares of the residuals the leastCorrect
Explanation. The method of least squares chooses \( a \) and \( b \) so that \( \sum (y_i - \hat{y}_i)^2 \), the sum of the squares of the residuals, is the least; the sum of the residuals is then zero.
Q12
Consider the following statements about regression coefficients:
1. They are independent of the change of origin.
2. They are independent of the change of scale.
3. Both regression coefficients have the same sign.
Which of these are correct?
- A. 1 and 2 only
- B. 1 and 3 onlyCorrect
- C. 2 and 3 only
- D. 1, 2 and 3
Explanation. Regression coefficients are independent of the change of origin but not of scale, so statement 2 is wrong. Since \( r = \sqrt{b_{XY} \times b_{YX}} \), both coefficients must have the same sign.
Q13
A regression equation of \( Y \) on \( X \) is fitted for \( x \) values ranging from 3.6 to 10.7. According to the book, it can be used to estimate \( Y \)
- A. for any value of \( x \), however large
- B. only for values of \( x \) outside 3.6 to 10.7 (extrapolation)
- C. only for values of \( x \) within 3.6 to 10.7 (interpolation)Correct
- D. only at \( x = \bar{x} \)
Explanation. The fitted equation can be used for prediction only for values of the regressor within its range, that is, interpolation. Results obtained by extrapolation cannot be interpreted.
Q14
Assertion (A): The correlation coefficient is symmetric in \( x \) and \( y \), but regression is not.
Reason (R): \( r_{XY} = r_{YX} \), while \( b_{XY} \) and \( b_{YX} \) have different meanings and interpretations.
- A. Both A and R are true and R is the correct explanation of ACorrect
- B. Both A and R are true but R is not the correct explanation of A
- C. A is true but R is false
- D. A is false but R is true
Explanation. In the comparison of correlation and regression, correlation is symmetric since \( r_{XY} = r_{YX} \); regression is not symmetric, because \( b_{XY} \) and \( b_{YX} \) differ in meaning. R explains A.
Q15
In fitting the simple linear regression \( Y = a + bX + e \), the error term \( e \) is assumed to be a random variable with
- A. mean one and constant variance
- B. mean zero and increasing variance
- C. mean equal to \( b \) and variance zero
- D. mean zero and constant varianceCorrect
Explanation. The general assumptions are that the relationship between \( Y \) and \( X \) is linear, the error term \( e \) is a random variable with mean zero, and \( e \) has constant variance.