CAIE FP2 2012 June — Question 11 OR

Exam BoardCAIE
ModuleFP2 (Further Pure Mathematics 2)
Year2012
SessionJune
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicLinear regression
TypeRelate two regression lines
DifficultyChallenging +1.2 This question tests standard relationships between regression lines and correlation coefficients (r² = product of gradients) plus using properties that regression lines pass through means. While it requires knowledge of multiple statistical concepts and algebraic manipulation, the techniques are well-established formulas from the Further Maths Statistics syllabus with straightforward application once the relationships are recalled.
Spec5.08c Pearson: measure of straight-line fit5.08d Hypothesis test: Pearson correlation5.09c Calculate regression line

For a random sample of 5 pairs of values of \(x\) and \(y\), the equations of the regression lines of \(y\) on \(x\) and \(x\) on \(y\) are respectively $$y = - 0.5 x + 5 \quad \text { and } \quad x = - 1.2 y + 7.6$$ Find the value of the product moment correlation coefficient for this sample. Test, at the \(5 \%\) significance level, whether the population product moment correlation coefficient differs from zero. The following table shows the sample data.
\(x\)1255\(p\)
\(y\)5342\(q\)
Find the values of \(p\) and \(q\).

Question 11(a):
AnswerMarks Guidance
Answer/WorkingMarks Guidance
\(R_C \cdot 3a = kW \cdot 4a\cos\theta + W \cdot 2a\cos\theta\)M1 A1 Take moments about \(A\) for rod to find \(R_C\)
\(R_C = (2k+1)(2W/5)\)A1
\(F_A = R_C\sin\theta = (2k+1)(8W/25)\)M1 A1 Resolve horizontal forces on rod to find \(F_A\)
\(R_A = W + kW - R_C\cos\theta = (13k+19)(W/25)\)M1 A1 Resolve vertical forces on rod to find \(R_A\)
*OR:* Take moments about \(C\) for rod:
\(F_A \cdot 3a\sin\theta - R_A \cdot 3a\cos\theta = kWa\cos\theta - Wa\cos\theta\)(M1 A1)
\(4F_A - 3R_A = (k-1)W\)(A1)
Resolve forces parallel to rod: \(F_A\cos\theta + R_A\sin\theta = kW\sin\theta + W\sin\theta\)(M1)
\(3F_A + 4R_A = 4(k+1)W\)(A1)
Combine to find \(F_A\), \(R_A\): \(F_A = (2k+1)(8W/25)\) and \(R_A = (13k+19)(W/25)\)(M1 A1)
Use \(F_A \le \mu R_A\) to find \(\mu_{\min}\): \(\mu_{\min} = 8(2k+1)/(13k+19)\) A.G.M1 A1 (No use of inequality loses A1)
\(160k + 80 \le (\text{or} =)\ 117k + 171\)M1 Find bound on \(k\) (allow use of equality)
Bound \(= 91/43\) or \(2.12\)A1
\(k \le 91/43\) or \(2.12\)A1 Correct use of inequality
Total: 12 marks
Question 11(b):
Find correlation coefficient \(r\):
AnswerMarks Guidance
\(r^2 = (-0.5)(-1.2) = 0.6 \text{ or } 0.775^2\)M1 A1
\(r = -0.775\)*A1 3 marks
State both hypotheses (B0 for \(r\)...):
AnswerMarks
\(H_0: \rho = 0,\ H_1: \rho \neq 0\)B1
State or use correct tabular one-tail \(r\) value:
AnswerMarks
\(r_{5,\ 2.5\%} = 0.878\)*B1
Valid method for reaching conclusion:
AnswerMarks Guidance
Accept \(H_0\) if \(r <\) tabular value
Correct conclusion (AEF, dep *A1, *B1):
AnswerMarks Guidance
Coeff. does not differ from zeroA1 4 marks
EITHER: Find two indep. eqns for \(p, q\):
AnswerMarks
\(14 + q = (-0.5)(13 + p) + 5 \times 5\)M1 A1
\(or\ (14 + q)/5 = (-0.5)(13 + p)/5 + 5\)
AnswerMarks
\(13 + p = (-1.2)(14 + q) + 5 \times 7.6\)
\(or\ (13 + p)/5 = (-1.2)(14 + q)/5 + 7.6\)A1
Combine to find \(p, q\):
AnswerMarks
\(p = 7,\ q = 1\)A1, A1
OR: Combine regression lines to find \(\bar{x},\ \bar{y}\):
AnswerMarks
\(\bar{x} = 4;\ \bar{y} = 3\)M1 A1; A1
Use sample data to find \(p, q\):
AnswerMarks Guidance
\(p = 7,\ q = 1\)A1, A1 5 marks
Total: [12]
## Question 11(a):

| Answer/Working | Marks | Guidance |
|---|---|---|
| $R_C \cdot 3a = kW \cdot 4a\cos\theta + W \cdot 2a\cos\theta$ | M1 A1 | Take moments about $A$ for rod to find $R_C$ |
| $R_C = (2k+1)(2W/5)$ | A1 | |
| $F_A = R_C\sin\theta = (2k+1)(8W/25)$ | M1 A1 | Resolve horizontal forces on rod to find $F_A$ |
| $R_A = W + kW - R_C\cos\theta = (13k+19)(W/25)$ | M1 A1 | Resolve vertical forces on rod to find $R_A$ |
| *OR:* Take moments about $C$ for rod: | | |
| $F_A \cdot 3a\sin\theta - R_A \cdot 3a\cos\theta = kWa\cos\theta - Wa\cos\theta$ | (M1 A1) | |
| $4F_A - 3R_A = (k-1)W$ | (A1) | |
| Resolve forces parallel to rod: $F_A\cos\theta + R_A\sin\theta = kW\sin\theta + W\sin\theta$ | (M1) | |
| $3F_A + 4R_A = 4(k+1)W$ | (A1) | |
| Combine to find $F_A$, $R_A$: $F_A = (2k+1)(8W/25)$ and $R_A = (13k+19)(W/25)$ | (M1 A1) | |
| Use $F_A \le \mu R_A$ to find $\mu_{\min}$: $\mu_{\min} = 8(2k+1)/(13k+19)$ **A.G.** | M1 A1 | (No use of inequality loses A1) |
| $160k + 80 \le (\text{or} =)\ 117k + 171$ | M1 | Find bound on $k$ (allow use of equality) |
| Bound $= 91/43$ or $2.12$ | A1 | |
| $k \le 91/43$ or $2.12$ | A1 | Correct use of inequality |

**Total: 12 marks**

## Question 11(b):

**Find correlation coefficient $r$:**

$r^2 = (-0.5)(-1.2) = 0.6 \text{ or } 0.775^2$ | M1 A1 |

$r = -0.775$ | *A1 | 3 marks

---

**State both hypotheses (B0 for $r$...):**

$H_0: \rho = 0,\ H_1: \rho \neq 0$ | B1 |

---

**State or use correct tabular one-tail $r$ value:**

$r_{5,\ 2.5\%} = 0.878$ | *B1 |

---

**Valid method for reaching conclusion:**

Accept $H_0$ if $|r| <$ tabular value | M1 |

---

**Correct conclusion (AEF, dep *A1, *B1):**

Coeff. does not differ from zero | A1 | 4 marks

---

**EITHER: Find two indep. eqns for $p, q$:**

$14 + q = (-0.5)(13 + p) + 5 \times 5$ | M1 A1 |

$or\ (14 + q)/5 = (-0.5)(13 + p)/5 + 5$

$13 + p = (-1.2)(14 + q) + 5 \times 7.6$ | |

$or\ (13 + p)/5 = (-1.2)(14 + q)/5 + 7.6$ | A1 |

---

**Combine to find $p, q$:**

$p = 7,\ q = 1$ | A1, A1 |

---

**OR: Combine regression lines to find $\bar{x},\ \bar{y}$:**

$\bar{x} = 4;\ \bar{y} = 3$ | M1 A1; A1 |

**Use sample data to find $p, q$:**

$p = 7,\ q = 1$ | A1, A1 | 5 marks

---

**Total: [12]**
For a random sample of 5 pairs of values of $x$ and $y$, the equations of the regression lines of $y$ on $x$ and $x$ on $y$ are respectively

$$y = - 0.5 x + 5 \quad \text { and } \quad x = - 1.2 y + 7.6$$

Find the value of the product moment correlation coefficient for this sample.

Test, at the $5 \%$ significance level, whether the population product moment correlation coefficient differs from zero.

The following table shows the sample data.

\begin{center}
\begin{tabular}{ | l | l | l | l | l | l | }
\hline
$x$ & 1 & 2 & 5 & 5 & $p$ \\
\hline
$y$ & 5 & 3 & 4 & 2 & $q$ \\
\hline
\end{tabular}
\end{center}

Find the values of $p$ and $q$.

\hfill \mbox{\textit{CAIE FP2 2012 Q11 OR}}