Edexcel S4 2018 June — Question 6 19 marks

Exam BoardEdexcel
ModuleS4 (Statistics 4)
Year2018
SessionJune
Marks19
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicDiscrete Random Variables
TypeShowing estimator is unbiased
DifficultyChallenging +1.2 This is a multi-part S4 question requiring integration for moments, variance calculation, and unbiased estimator theory. While it involves several steps and statistical concepts (unbiasedness, consistency, minimum variance), each part follows standard techniques taught in Further Maths Statistics. The integration is straightforward, and the optimization in part (d) uses standard Lagrange multipliers or substitution methods. More routine than challenging for S4 students.
Spec5.03a Continuous random variables: pdf and cdf5.03b Solve problems: using pdf5.03c Calculate mean/variance: by integration5.05b Unbiased estimates: of population mean and variance

  1. The continuous random variable \(X\) has probability density function \(\mathrm { f } ( x )\)
$$f ( x ) = \left\{ \begin{array} { c c } \frac { x } { 2 \theta ^ { 2 } } & 0 \leqslant x \leqslant 2 \theta \\ 0 & \text { otherwise } \end{array} \right.$$ where \(\theta\) is a constant.
  1. Use integration to show that \(\mathrm { E } \left( X ^ { N } \right) = \frac { 2 ^ { N + 1 } } { N + 2 } \theta ^ { N }\)
  2. Hence
    1. write down an expression for \(\mathrm { E } ( X )\) in terms of \(\theta\)
    2. find \(\operatorname { Var } ( X )\) in terms of \(\theta\) A random sample \(X _ { 1 } , X _ { 2 } , \ldots , X _ { n }\) where \(n \geqslant 2\) is taken to estimate the value of \(\theta\) The random variable \(S _ { 1 } = q \bar { X }\) is an unbiased estimator of \(\theta\)
  3. Write down the value of \(q\) and show that \(S _ { 1 }\) is a consistent estimator of \(\theta\) The continuous random variable \(Y\) is independent of \(X\) and is uniformly distributed over the interval \(\left[ 0 , \frac { 2 \theta } { 3 } \right]\), where \(\theta\) is the same unknown constant as in \(\mathrm { f } ( x )\). The random variable \(S _ { 2 } = a X + b Y\) is an unbiased estimator of \(\theta\) and is based on one observation of \(X\) and one observation of \(Y\).
  4. Find the value of \(a\) and the value of \(b\) for which \(S _ { 2 }\) has minimum variance.
  5. Show that the minimum variance of \(S _ { 2 }\) is \(\frac { \theta ^ { 2 } } { 11 }\)
  6. Explain which of \(S _ { 1 }\) or \(S _ { 2 }\) is the better estimator for \(\theta\)

Question 6:
Part (a)
AnswerMarks Guidance
AnswerMark Guidance
\(E(X^N) = \int_0^{2\theta} \frac{x^{N+1}}{2\theta^2}dx\)M1 Attempting to integrate \(\frac{x^{N+1}}{2\theta^2}\); \(x^{N+1} \to x^{N+2}\); condone missing limits
\(= \left[\frac{x^{N+2}}{2(N+2)\theta^2}\right]_0^{2\theta}\)A1 Correct integration
\(= \frac{(2\theta)^{N+2}}{2(N+2)\theta^2} = \frac{2^{N+1}}{N+2}\theta^N\)A1cso (3) Must see substitution of \(2\theta\)
Part (b)
AnswerMarks Guidance
AnswerMark Guidance
\(E(X) = \frac{4\theta}{3}\)B1 Must have \(E(X) = \); allow their \(E(X)\) if one has been given otherwise must be correct
\(\text{Var}(X) = 2\theta^2 - \left(\frac{\text{"4}\theta\text{"} }{3}\right)^2 = \frac{2\theta^2}{9}\)M1A1 (3) A1 must be using part (a); do not allow if integrated from scratch
Part (c)
AnswerMarks Guidance
AnswerMark Guidance
\(q = \frac{3}{4}\)B1
\(\text{Var}(S_1) = \frac{9}{16} \times \frac{\text{"}\frac{2\theta^2}{9}\text{"}}{n} = \frac{\theta^2}{8n}\)M1, A1cso M1 for \(\frac{9}{16}\times\frac{\text{their Var}(X)}{n}\)
as \(n\to\infty\), \(\text{Var}(S) \to 0\); since it is unbiased, it is a consistent estimator(3) A1 cso and for \(n\to\infty\), \(\text{Var}(S)\to 0\); since it is unbiased it is a consistent estimator
Part (d)
AnswerMarks Guidance
AnswerMark Guidance
\(E(S_2) = a\times\frac{\text{"4}\theta\text{"}}{3} + b\times\frac{\theta}{3}\)M1 M1 for \(a\times\text{their }E(X) + b\times\frac{\theta}{3}\)
\(a\times\frac{4\theta}{3} + b\times\frac{\theta}{3} = \theta\) or \(4a + b = 3\)A1 Correct equation with no \(\theta\)
\(\text{Var}(S_2) = a^2\times\frac{\text{"}\frac{2\theta^2}{9}\text{"}}{} + b^2\times\frac{\theta^2}{27}\)M1 M1 \(a^2\times\text{their Var}(X) + b^2\times\frac{\theta^2}{27}\)
\(\text{Var}(S_2) = a^2\times\frac{2\theta^2}{9} + (3-4a)^2\times\frac{\theta^2}{27}\) or \(\text{Var}(S_2) = \left(\frac{3-b}{4}\right)^2\times\frac{2\theta^2}{9} + b^2\times\frac{\theta^2}{27}\)M1 M1 subst in for \(a\) or \(b\)
\(\frac{d\text{Var}(S_2)}{da} = \frac{4a\theta^2}{9} - \frac{8(3-4a)\theta^2}{27} = 0\) or \(\frac{-(3-b)\theta^2}{36} + \frac{2b\theta^2}{27} = 0\)M1 M1 differentiating with respect to \(a\) or \(b\)
\(\frac{44a}{27} = \frac{24}{27}\) or \(\frac{11}{108}b = \frac{1}{12}\)M1 M1 putting \(d\text{Var}/da = 0\) and solving
\(a = \frac{6}{11}\), \(b = \frac{9}{11}\)A1 (7) Allow awrt 0.545 and awrt 0.818
Part (e)
AnswerMarks Guidance
AnswerMark Guidance
\(\text{Var}(S_2) = \left(\frac{\text{"6"}}{11}\right)^2 \times \frac{2\theta^2}{9} + \left(\frac{\text{"9"}}{11}\right)^2 \times \frac{\theta^2}{27}\)M1 M1 subst \(a\) and \(b\) in to find \(\text{Var}(S_2)\)
\(\text{Var}(S_2) = \frac{\theta^2}{11}\)(1)
Part (f)
AnswerMarks Guidance
AnswerMark Guidance
\(S_1\) is the better estimator when \(\frac{\theta^2}{8n} < \frac{\theta^2}{11} \Rightarrow n > \frac{11}{8}\)M1 M1 for reason \(\frac{\theta^2}{8n} < \frac{\theta^2}{11} \Rightarrow n > \frac{11}{8}\) or \(\text{Var}(S_1) \leq \frac{\theta^2}{16} < \frac{\theta^2}{11}\)
\(S_2\) is the better estimator when \(n < \frac{11}{8}\)
Therefore \(S_1\) is the better estimator since \(n \geq 2\)A1cso (2) Correct selection
Total: 19
The image appears to be essentially blank/white, containing only the Pearson Education Limited copyright notice at the bottom. There is no mark scheme content visible on this page to extract.
## Question 6:

### Part (a)
| Answer | Mark | Guidance |
|--------|------|----------|
| $E(X^N) = \int_0^{2\theta} \frac{x^{N+1}}{2\theta^2}dx$ | M1 | Attempting to integrate $\frac{x^{N+1}}{2\theta^2}$; $x^{N+1} \to x^{N+2}$; condone missing limits |
| $= \left[\frac{x^{N+2}}{2(N+2)\theta^2}\right]_0^{2\theta}$ | A1 | Correct integration |
| $= \frac{(2\theta)^{N+2}}{2(N+2)\theta^2} = \frac{2^{N+1}}{N+2}\theta^N$ | A1cso (3) | Must see substitution of $2\theta$ |

### Part (b)
| Answer | Mark | Guidance |
|--------|------|----------|
| $E(X) = \frac{4\theta}{3}$ | B1 | Must have $E(X) = $; allow their $E(X)$ if one has been given otherwise must be correct |
| $\text{Var}(X) = 2\theta^2 - \left(\frac{\text{"4}\theta\text{"} }{3}\right)^2 = \frac{2\theta^2}{9}$ | M1A1 (3) | A1 must be using part (a); do not allow if integrated from scratch |

### Part (c)
| Answer | Mark | Guidance |
|--------|------|----------|
| $q = \frac{3}{4}$ | B1 | |
| $\text{Var}(S_1) = \frac{9}{16} \times \frac{\text{"}\frac{2\theta^2}{9}\text{"}}{n} = \frac{\theta^2}{8n}$ | M1, A1cso | M1 for $\frac{9}{16}\times\frac{\text{their Var}(X)}{n}$ |
| as $n\to\infty$, $\text{Var}(S) \to 0$; since it is unbiased, it is a consistent estimator | (3) | A1 cso and for $n\to\infty$, $\text{Var}(S)\to 0$; since it is unbiased it is a consistent estimator |

### Part (d)
| Answer | Mark | Guidance |
|--------|------|----------|
| $E(S_2) = a\times\frac{\text{"4}\theta\text{"}}{3} + b\times\frac{\theta}{3}$ | M1 | M1 for $a\times\text{their }E(X) + b\times\frac{\theta}{3}$ |
| $a\times\frac{4\theta}{3} + b\times\frac{\theta}{3} = \theta$ or $4a + b = 3$ | A1 | Correct equation with no $\theta$ |
| $\text{Var}(S_2) = a^2\times\frac{\text{"}\frac{2\theta^2}{9}\text{"}}{} + b^2\times\frac{\theta^2}{27}$ | M1 | M1 $a^2\times\text{their Var}(X) + b^2\times\frac{\theta^2}{27}$ |
| $\text{Var}(S_2) = a^2\times\frac{2\theta^2}{9} + (3-4a)^2\times\frac{\theta^2}{27}$ or $\text{Var}(S_2) = \left(\frac{3-b}{4}\right)^2\times\frac{2\theta^2}{9} + b^2\times\frac{\theta^2}{27}$ | M1 | M1 subst in for $a$ or $b$ |
| $\frac{d\text{Var}(S_2)}{da} = \frac{4a\theta^2}{9} - \frac{8(3-4a)\theta^2}{27} = 0$ or $\frac{-(3-b)\theta^2}{36} + \frac{2b\theta^2}{27} = 0$ | M1 | M1 differentiating with respect to $a$ or $b$ |
| $\frac{44a}{27} = \frac{24}{27}$ or $\frac{11}{108}b = \frac{1}{12}$ | M1 | M1 putting $d\text{Var}/da = 0$ and solving |
| $a = \frac{6}{11}$, $b = \frac{9}{11}$ | A1 (7) | Allow awrt 0.545 and awrt 0.818 |

### Part (e)
| Answer | Mark | Guidance |
|--------|------|----------|
| $\text{Var}(S_2) = \left(\frac{\text{"6"}}{11}\right)^2 \times \frac{2\theta^2}{9} + \left(\frac{\text{"9"}}{11}\right)^2 \times \frac{\theta^2}{27}$ | M1 | M1 subst $a$ and $b$ in to find $\text{Var}(S_2)$ |
| $\text{Var}(S_2) = \frac{\theta^2}{11}$ | (1) | |

### Part (f)
| Answer | Mark | Guidance |
|--------|------|----------|
| $S_1$ is the better estimator when $\frac{\theta^2}{8n} < \frac{\theta^2}{11} \Rightarrow n > \frac{11}{8}$ | M1 | M1 for reason $\frac{\theta^2}{8n} < \frac{\theta^2}{11} \Rightarrow n > \frac{11}{8}$ or $\text{Var}(S_1) \leq \frac{\theta^2}{16} < \frac{\theta^2}{11}$ |
| $S_2$ is the better estimator when $n < \frac{11}{8}$ | | |
| Therefore $S_1$ is the better estimator since $n \geq 2$ | A1cso (2) | Correct selection |

**Total: 19**

The image appears to be essentially blank/white, containing only the Pearson Education Limited copyright notice at the bottom. There is no mark scheme content visible on this page to extract.
\begin{enumerate}
  \item The continuous random variable $X$ has probability density function $\mathrm { f } ( x )$
\end{enumerate}

$$f ( x ) = \left\{ \begin{array} { c c } 
\frac { x } { 2 \theta ^ { 2 } } & 0 \leqslant x \leqslant 2 \theta \\
0 & \text { otherwise }
\end{array} \right.$$

where $\theta$ is a constant.\\
(a) Use integration to show that $\mathrm { E } \left( X ^ { N } \right) = \frac { 2 ^ { N + 1 } } { N + 2 } \theta ^ { N }$\\
(b) Hence\\
(i) write down an expression for $\mathrm { E } ( X )$ in terms of $\theta$\\
(ii) find $\operatorname { Var } ( X )$ in terms of $\theta$

A random sample $X _ { 1 } , X _ { 2 } , \ldots , X _ { n }$ where $n \geqslant 2$ is taken to estimate the value of $\theta$ The random variable $S _ { 1 } = q \bar { X }$ is an unbiased estimator of $\theta$\\
(c) Write down the value of $q$ and show that $S _ { 1 }$ is a consistent estimator of $\theta$

The continuous random variable $Y$ is independent of $X$ and is uniformly distributed over the interval $\left[ 0 , \frac { 2 \theta } { 3 } \right]$, where $\theta$ is the same unknown constant as in $\mathrm { f } ( x )$.

The random variable $S _ { 2 } = a X + b Y$ is an unbiased estimator of $\theta$ and is based on one observation of $X$ and one observation of $Y$.\\
(d) Find the value of $a$ and the value of $b$ for which $S _ { 2 }$ has minimum variance.\\
(e) Show that the minimum variance of $S _ { 2 }$ is $\frac { \theta ^ { 2 } } { 11 }$\\
(f) Explain which of $S _ { 1 }$ or $S _ { 2 }$ is the better estimator for $\theta$

\hfill \mbox{\textit{Edexcel S4 2018 Q6 [19]}}