Edexcel S2 2012 January — Question 4 16 marks

Exam BoardEdexcel
ModuleS2 (Statistics 2)
Year2012
SessionJanuary
Marks16
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicPoisson distribution
TypePoisson approximation justification or comparison
DifficultyModerate -0.3 This is a straightforward application of the Poisson distribution with standard conditions recall and routine calculations. Part (a)-(b) test basic knowledge of when Poisson applies, (c)-(d) are direct probability calculations with given parameters, and (e) requires the standard normal approximation technique. All steps are textbook procedures with no novel problem-solving required, making it slightly easier than average.
Spec2.04d Normal approximation to binomial5.02i Poisson distribution: random events model5.02j Poisson formula: P(X=x) = e^(-lambda)*lambda^x/x!5.02k Calculate Poisson probabilities

4. A website receives hits at a rate of 300 per hour.
  1. State a distribution that is suitable to model the number of hits obtained during a 1 minute interval.
  2. State two reasons for your answer to part (a). Find the probability of
  3. 10 hits in a given minute,
  4. at least 15 hits in 2 minutes. The website will go down if there are more than 70 hits in 10 minutes.
  5. Using a suitable approximation, find the probability that the website will go down in a particular 10 minute interval.

Question 4:
Part (a):
AnswerMarks Guidance
Answer/WorkingMark Guidance
PoissonB1
Part (b):
AnswerMarks Guidance
Answer/WorkingMark Guidance
Hits occur singly in timeB1 Any one of the 3 statements — no context required. Must be constant (mean) rate, not constant probability or constant mean
Hits are independent or Hits occur randomlyB1 A different statement with context of hits. NB random and independent are the same statement. If only one mark awarded, give 1st B1. Never award B0B1
Hits occur at a constant rate
Part (c):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(X \sim Po(5)\)B1 Writing or using \(Po(5)\)
\(P(X=10) = P(X \leq 10) - P(X \leq 9)\) or \(\frac{e^{-5}5^{10}}{10!}\)M1
\(= 0.9863 - 0.9682 = 0.0181\)A1 awrt 0.0181
Part (d):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(X \sim Po(10)\)B1 Writing or using \(Po(10)\)
\(P(X \geq 15) = 1 - P(X \leq 14)\)M1 Writing or using \(1 - P(X \leq 14)\)
\(= 1 - 0.9165 = 0.0835\)A1 awrt 0.0835
Part (e):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(X \sim Po(50)\), Approximated by \(N(50, 50)\)B1B1 1st B1 for normal approximation. 2nd B1 for correct mean and sd
\(P(X > 70) = P\!\left(Z > \frac{70.5 - 50}{\sqrt{50}}\right)\)M1M1 1st M1 for attempting continuity correction \((71 \pm 0.5)\). 2nd M1 for standardising using their mean and sd; allow \([69.5, 70, 70.5, 71, 71.5]\), allow \(\pm z\)
\(= P(Z > 2.899\ldots)\)A1 \(z = \pm\) awrt 2.9 or better
\(= 1 - 0.9981\)M1 For \(1 -\) tables value
\(= 0.0019\)A1 awrt 0.0019
SC using \(P(X < 70.5/71.5) - P(X \leq 69.5/70.5)\) can get B1B1 M0M1A0 M0A0
# Question 4:

## Part (a):

| Answer/Working | Mark | Guidance |
|---|---|---|
| Poisson | B1 | |

## Part (b):

| Answer/Working | Mark | Guidance |
|---|---|---|
| Hits occur **singly** in time | B1 | Any one of the 3 statements — no context required. Must be constant (mean) rate, not constant probability or constant mean |
| Hits are **independent** or Hits occur **randomly** | B1 | A different statement with context of **hits**. NB random and independent are the same statement. If only one mark awarded, give 1st B1. Never award B0B1 |
| Hits occur at a **constant rate** | | |

## Part (c):

| Answer/Working | Mark | Guidance |
|---|---|---|
| $X \sim Po(5)$ | B1 | Writing or using $Po(5)$ |
| $P(X=10) = P(X \leq 10) - P(X \leq 9)$ or $\frac{e^{-5}5^{10}}{10!}$ | M1 | |
| $= 0.9863 - 0.9682 = 0.0181$ | A1 | awrt 0.0181 |

## Part (d):

| Answer/Working | Mark | Guidance |
|---|---|---|
| $X \sim Po(10)$ | B1 | Writing or using $Po(10)$ |
| $P(X \geq 15) = 1 - P(X \leq 14)$ | M1 | Writing or using $1 - P(X \leq 14)$ |
| $= 1 - 0.9165 = 0.0835$ | A1 | awrt 0.0835 |

## Part (e):

| Answer/Working | Mark | Guidance |
|---|---|---|
| $X \sim Po(50)$, Approximated by $N(50, 50)$ | B1B1 | 1st B1 for normal approximation. 2nd B1 for correct mean and sd |
| $P(X > 70) = P\!\left(Z > \frac{70.5 - 50}{\sqrt{50}}\right)$ | M1M1 | 1st M1 for attempting continuity correction $(71 \pm 0.5)$. 2nd M1 for standardising using their mean and sd; allow $[69.5, 70, 70.5, 71, 71.5]$, allow $\pm z$ |
| $= P(Z > 2.899\ldots)$ | A1 | $z = \pm$ awrt 2.9 or better |
| $= 1 - 0.9981$ | M1 | For $1 -$ tables value |
| $= 0.0019$ | A1 | awrt 0.0019 |
| SC using $P(X < 70.5/71.5) - P(X \leq 69.5/70.5)$ can get B1B1 M0M1A0 M0A0 | | |

---
4. A website receives hits at a rate of 300 per hour.
\begin{enumerate}[label=(\alph*)]
\item State a distribution that is suitable to model the number of hits obtained during a 1 minute interval.
\item State two reasons for your answer to part (a).

Find the probability of
\item 10 hits in a given minute,
\item at least 15 hits in 2 minutes.

The website will go down if there are more than 70 hits in 10 minutes.
\item Using a suitable approximation, find the probability that the website will go down in a particular 10 minute interval.
\end{enumerate}

\hfill \mbox{\textit{Edexcel S2 2012 Q4 [16]}}