‹ Back to the paper
Question 21(ii)
4 marks
MathematicsISC 2026 Mathematics Paper 1Case basedLinear Regression
Answer the following on regression and correlation.
If the regression lines of a bivariate distribution are 4x - 5y + 33 = 0 and 20x - 9y - 107 = 0…
If the regression lines of a bivariate distribution are $4x - 5y + 33 = 0$ and $20x - 9y - 107 = 0$, then
(a)[1]
Calculate the arithmetic mean of $x$ and $y$.
(b)[2]
Estimate the value of $x$ when $y = 7$.
(c)[1]
Find the variance of $y$ when $\sigma_x = 3$.
Answer
Answer (a)
The means satisfy both lines: $4x - 5y + 33 = 0$ and $20x - 9y - 107 = 0$.
From the first, $x = \frac{5y - 33}{4}$. Substituting: $25y - 165 - 9y - 107 = 0 \Rightarrow 16y = 272 \Rightarrow y = 17$, then $x = \frac{85-33}{4} = 13$.
Hence $\bar{x} = 13$ and $\bar{y} = 17$.
Final answer: $\bar{x} = 13,\ \bar{y} = 17$
Answer (b)
Taking $4x - 5y + 33 = 0$ as $y$ on $x$: $b_{yx} = \frac{4}{5} = 0.8$; and $20x - 9y - 107 = 0$ as $x$ on $y$: $b_{xy} = \frac{9}{20} = 0.45$. Then $r^2 = 0.8 \times 0.45 = 0.36 < 1$, so this choice is valid (the other choice gives $b_{xy}b_{yx} > 1$).
Line of $x$ on $y$: $x = \frac{9y + 107}{20}$. At $y = 7$: $x = \frac{63 + 107}{20} = 8.5$.
Final answer: 8.5
Answer (c)
$r = +\sqrt{0.36} = 0.6$ (both coefficients are positive).
$b_{yx} = r\frac{\sigma_y}{\sigma_x} \Rightarrow 0.8 = 0.6\cdot\frac{\sigma_y}{3} \Rightarrow \sigma_y = 4$.
Variance of $y$ $= \sigma_y^2 = 16$.
Final answer: 16