| Exam Board | CAIE |
|---|---|
| Module | FP2 (Further Pure Mathematics 2) |
| Year | 2019 |
| Session | June |
| Marks | 11 |
| Paper | Download PDF ↗ |
| Mark scheme | Download PDF ↗ |
| Topic | Linear regression |
| Type | Find unknown values from regression |
| Difficulty | Standard +0.3 This is a straightforward application of regression formulas with one algebraic twist. Students must use the property that regression lines pass through the mean point to find q, then calculate c and the correlation coefficient using standard formulas. While it requires careful algebraic manipulation and knowledge of regression theory, it's a routine multi-part question testing standard techniques without requiring novel insight or complex problem-solving. |
| Spec | 5.08a Pearson correlation: calculate pmcc5.08d Hypothesis test: Pearson correlation5.09a Dependent/independent variables5.09b Least squares regression: concepts5.09c Calculate regression line5.09d Linear coding: effect on regression |
| \(x\) | 3 | 4 | 4 | 6 | 8 |
| \(y\) | 5 | 7 | \(q\) | 6 | 7 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Marks | Guidance |
| \(\sum x=25\), \(\sum y=25+q\), \(\sum xy=135+4q\), \(\sum x^2=141\), \(\sum y^2=159+q^2\) | B1 | Find required values |
| \(S_{xy}=135+4q-25(25+q)/5=[10-q]\) (AEF); \([S_{xx}=141-25^2/5=16]\); \(S_{yy}=(159+q^2)-(25+q)^2/5\ [=34-10q+4q^2/5]\) | M1 A1 | (\(S_{xy}\), \(S_{xx}\), \(S_{yy}\) may be scaled by the same constant) |
| \(40-4q=170-50q+4q^2\), \(2q^2-23q+65=0\), \((2q-13)(q-5)=0\), \(q=5\) | M1 A1 | Equate gradient \(\frac{5}{4}\) in line of \(x\) on \(y\) to \(S_{xy}/S_{yy}\) and solve quadratic to find integer value of \(q\) |
| 5 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Marks | Guidance |
| \(c = 25/5 - (5/4)(25+q)/5 = -(5+q)/4\) | M1 A1 | Find \(c\) from \(\bar{x} - \frac{5}{4}\,\bar{y}\) |
| \(= -5/2\) or \(-2.5\) | A1 | |
| 3 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Marks | Guidance |
| \(r = S_{xy}/\sqrt{(S_{xx}S_{yy})} = 5/\sqrt{(16\times 4)}\) | M1 A1 | Find correlation coefficient \(r\) |
| \(= 5/8\) or \(0.625\) | A1 | |
| 3 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Marks | Guidance |
| \(\frac{1}{2}mu_P^2 = \frac{1}{2}m(21ag/2) + mga\quad [u_P^2 = \frac{25}{2}ag]\) | M1 A1 | \(u_P\) is speed of \(P\) at lowest point, \(v_Q\) is speed of \(Q\) immediately after collision. Apply conservation of energy at lowest point (A0 if no \(m\)) |
| \(\frac{1}{2}\cdot4mv_Q^2 = 4mag\) | M1 | Find speed \(v_Q\) at lowest point by conservation of energy (A0 if no \(m\)) |
| \(v_Q = \sqrt{2ag}\) or \(1.41\sqrt{ag}\) or \(4.47\sqrt{a}\) | A1 | |
| \(mu_P = [\pm]\,mv_P + 4mv_Q\) | M1 | Find \(v_P\) using conservation of momentum (\(m\) may be omitted) |
| \(v_P = [\pm](-5/\sqrt{2}+4\sqrt{2})\sqrt{ag}\) | A1 | |
| \( | v_P | = \frac{3}{\sqrt{2}}\sqrt{ag}\) or \(2.12\sqrt{ag}\) or \(0.671\sqrt{a}\) (AEF) |
| 7 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Marks | Guidance |
| \(V_P\) is speed of \(P\) when it loses contact: \(\frac{1}{2}mV_P^2 = \frac{1}{2}mv_P^2 - mga(1+\cos\alpha)\); \([V_P^2 = \frac{9}{2}ag - 2ga(1+\cos\alpha) = (\frac{5}{2}-2\cos\alpha)ag]\) | M1 A1 | Apply conservation of energy at \(D\) (A0 if no \(m\)) |
| \([R_D=]\,mV_P^2/a - mg\cos\alpha = 0\quad [V_P^2 = ag\cos\alpha]\) | M1 A1 | Apply \(F=ma\) radially at \(D\) with reaction \(= 0\) |
| \((\frac{5}{2}-2\cos\alpha)ag = ag\cos\alpha\), \(\cos\alpha = 5/6\) or \(0.833\) | A1 | Combine to find \(\cos\alpha\) |
| 5 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| \(t_{s_A} / \sqrt{8} = \frac{1}{2}(16.7 - 13.5) [= 1.6]\) | M1 | Relate \(s_A\) to semi-width of confidence interval |
| \(t_{7, 0.975} = 2.365\) (to 3 s.f.) | A1 | State or use correct tabular \(t\) value |
| \([s_A = \sqrt{8} \times 1.6 / 2.365 = 1.9135]\), \(s_A^2 = 3.66[16]\) | A1 | Hence find unbiased estimate of \(A\)'s population variance |
| 3 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| \(H_0: \mu_A = \mu_B\), \(H_1: \mu_A > \mu_B\) (AEF) | B1 | State hypotheses (B0 for \(\bar{x}\) ...) |
| \([\bar{x}_A = 15.1]\), \(\bar{x}_B = 85.2/6 = 14.2\) | B1 | Find sample mean for \(B\) |
| \(s_B^2 = (1221.06 - 85.2^2/6)/5 = 561/250\) or \(2.244\) or \(1.498^2\) (all to 3 s.f.) | M1 | Estimate or imply population variance for \(B\) (allow biased here: \(1.87\) or \(1.367^2\)) |
| \(s^2 = (7s_A^2 + 5s_B^2)/12 = 3.0709\) or \(1.752^2\) | M1 A1 | Estimate (pooled) common variance (\(s_B^2\) not needed explicitly) |
| \(t_{12, 0.95} = 1.782\) | B1 | State or use correct tabular \(t\) value |
| \([-]\, t = (\bar{x}_A - \bar{x}_B) / (s\sqrt{1/8 + 1/6}) = 0.951\), \(t < 1.78\) so [accept \(H_0\)] | M1 A1 | Find value of \(t\) (or can compare \(\bar{x}_A - \bar{x}_B = 0.9\) with \(1.69\)); Correct conclusion |
| mean mass of \(B\) not less than mean mass of \(A\) (AEF) | B1 | |
| 9 |
## Question 10(i):
| Answer | Marks | Guidance |
|--------|-------|----------|
| $\sum x=25$, $\sum y=25+q$, $\sum xy=135+4q$, $\sum x^2=141$, $\sum y^2=159+q^2$ | B1 | Find required values |
| $S_{xy}=135+4q-25(25+q)/5=[10-q]$ (AEF); $[S_{xx}=141-25^2/5=16]$; $S_{yy}=(159+q^2)-(25+q)^2/5\ [=34-10q+4q^2/5]$ | M1 A1 | ($S_{xy}$, $S_{xx}$, $S_{yy}$ may be scaled by the same constant) |
| $40-4q=170-50q+4q^2$, $2q^2-23q+65=0$, $(2q-13)(q-5)=0$, $q=5$ | M1 A1 | Equate gradient $\frac{5}{4}$ in line of $x$ on $y$ to $S_{xy}/S_{yy}$ and solve quadratic to find integer value of $q$ |
| | **5** | |
## Question 10(ii):
| Answer | Marks | Guidance |
|--------|-------|----------|
| $c = 25/5 - (5/4)(25+q)/5 = -(5+q)/4$ | M1 A1 | Find $c$ from $\bar{x} - \frac{5}{4}\,\bar{y}$ |
| $= -5/2$ or $-2.5$ | A1 | |
| | **3** | |
## Question 10(iii):
| Answer | Marks | Guidance |
|--------|-------|----------|
| $r = S_{xy}/\sqrt{(S_{xx}S_{yy})} = 5/\sqrt{(16\times 4)}$ | M1 A1 | Find correlation coefficient $r$ |
| $= 5/8$ or $0.625$ | A1 | |
| | **3** | |
## Question 11E(i):
| Answer | Marks | Guidance |
|--------|-------|----------|
| $\frac{1}{2}mu_P^2 = \frac{1}{2}m(21ag/2) + mga\quad [u_P^2 = \frac{25}{2}ag]$ | M1 A1 | $u_P$ is speed of $P$ at lowest point, $v_Q$ is speed of $Q$ immediately after collision. Apply conservation of energy at lowest point (A0 if no $m$) |
| $\frac{1}{2}\cdot4mv_Q^2 = 4mag$ | M1 | Find speed $v_Q$ at lowest point by conservation of energy (A0 if no $m$) |
| $v_Q = \sqrt{2ag}$ or $1.41\sqrt{ag}$ or $4.47\sqrt{a}$ | A1 | |
| $mu_P = [\pm]\,mv_P + 4mv_Q$ | M1 | Find $v_P$ using conservation of momentum ($m$ may be omitted) |
| $v_P = [\pm](-5/\sqrt{2}+4\sqrt{2})\sqrt{ag}$ | A1 | |
| $|v_P| = \frac{3}{\sqrt{2}}\sqrt{ag}$ or $2.12\sqrt{ag}$ or $0.671\sqrt{a}$ (AEF) | A1 | Hence find speed of $P$ |
| | **7** | |
## Question 11E(ii):
| Answer | Marks | Guidance |
|--------|-------|----------|
| $V_P$ is speed of $P$ when it loses contact: $\frac{1}{2}mV_P^2 = \frac{1}{2}mv_P^2 - mga(1+\cos\alpha)$; $[V_P^2 = \frac{9}{2}ag - 2ga(1+\cos\alpha) = (\frac{5}{2}-2\cos\alpha)ag]$ | M1 A1 | Apply conservation of energy at $D$ (A0 if no $m$) |
| $[R_D=]\,mV_P^2/a - mg\cos\alpha = 0\quad [V_P^2 = ag\cos\alpha]$ | M1 A1 | Apply $F=ma$ radially at $D$ with reaction $= 0$ |
| $(\frac{5}{2}-2\cos\alpha)ag = ag\cos\alpha$, $\cos\alpha = 5/6$ or $0.833$ | A1 | Combine to find $\cos\alpha$ |
| | **5** | |
## Question 11O(i):
| Answer | Mark | Guidance |
|--------|------|----------|
| $t_{s_A} / \sqrt{8} = \frac{1}{2}(16.7 - 13.5) [= 1.6]$ | M1 | Relate $s_A$ to semi-width of confidence interval |
| $t_{7, 0.975} = 2.365$ (to 3 s.f.) | A1 | State or use correct tabular $t$ value |
| $[s_A = \sqrt{8} \times 1.6 / 2.365 = 1.9135]$, $s_A^2 = 3.66[16]$ | A1 | Hence find unbiased estimate of $A$'s population variance |
| | **3** | |
## Question 11O(ii):
| Answer | Mark | Guidance |
|--------|------|----------|
| $H_0: \mu_A = \mu_B$, $H_1: \mu_A > \mu_B$ (AEF) | B1 | State hypotheses (B0 for $\bar{x}$ ...) |
| $[\bar{x}_A = 15.1]$, $\bar{x}_B = 85.2/6 = 14.2$ | B1 | Find sample mean for $B$ |
| $s_B^2 = (1221.06 - 85.2^2/6)/5 = 561/250$ or $2.244$ or $1.498^2$ (all to 3 s.f.) | M1 | Estimate or imply population variance for $B$ (allow biased here: $1.87$ or $1.367^2$) |
| $s^2 = (7s_A^2 + 5s_B^2)/12 = 3.0709$ or $1.752^2$ | M1 A1 | Estimate (pooled) common variance ($s_B^2$ not needed explicitly) |
| $t_{12, 0.95} = 1.782$ | B1 | State or use correct tabular $t$ value |
| $[-]\, t = (\bar{x}_A - \bar{x}_B) / (s\sqrt{1/8 + 1/6}) = 0.951$, $t < 1.78$ so [accept $H_0$] | M1 A1 | Find value of $t$ (or can compare $\bar{x}_A - \bar{x}_B = 0.9$ with $1.69$); Correct conclusion |
| mean mass of $B$ not less than mean mass of $A$ (AEF) | B1 | |
| | **9** | |
**SC1:** Implicitly taking $s_A^2$, $s_B^2$ as unequal population variances (may also earn first B1 B1 M1):
$$z = (\bar{x}_A - \bar{x}_B) / \sqrt{(s_A^2/8 + s_B^2/6)} = 0.9/\sqrt{(0.8317)} = 0.987$$
$z < 1.645$ so
**DepSC1:** mean mass of $B$ not less than mean mass of $A$ (AEF); Comparison with $z_{0.95}$ and conclusion (FT on $z$); (can earn at most 5/9)
10 The values from a random sample of five pairs $( x , y )$ taken from a bivariate distribution are shown below.
\begin{center}
\begin{tabular}{ | l | l | l | l | l | l | }
\hline
$x$ & 3 & 4 & 4 & 6 & 8 \\
\hline
$y$ & 5 & 7 & $q$ & 6 & 7 \\
\hline
\end{tabular}
\end{center}
The equation of the regression line of $x$ on $y$ is given by $x = \frac { 5 } { 4 } y + c$.\\
(i) Given that $q$ is an integer, find its value.\\
(ii) Find the value of $c$.\\
(iii) Find the value of the product moment correlation coefficient.\\
\hfill \mbox{\textit{CAIE FP2 2019 Q10 [11]}}