| Exam Board | AQA |
|---|---|
| Module | D2 (Decision Mathematics 2) |
| Year | 2011 |
| Session | June |
| Marks | 15 |
| Paper | Download PDF ↗ |
| Mark scheme | Download PDF ↗ |
| Topic | Groups |
| Difficulty | Easy -1.8 This is a standard game theory question from Decision Mathematics involving routine application of minimax theorem and mixed strategy calculations. The methods are algorithmic (finding row minima/column maxima, setting up equations for mixed strategies) with no conceptual depth or novel problem-solving required. This is significantly easier than typical pure mathematics A-level questions. |
| Spec | 7.08a Pay-off matrix: zero-sum games7.08b Dominance: reduce pay-off matrix7.08c Pure strategies: play-safe strategies and stable solutions7.08d Nash equilibrium: identification and interpretation7.08e Mixed strategies: optimal strategy using equations or graphical method |
| Jerry | ||||
| \cline { 2 - 5 } | Strategy | A | B | C |
| Tom | I | - 4 | 5 | - 3 |
| \cline { 2 - 5 } | II | - 3 | - 2 | 8 |
| \cline { 2 - 5 } | III | - 7 | 6 | - 2 |
| Strategy | \(\mathbf { C } _ { \mathbf { 1 } }\) | \(\mathbf { C } _ { \mathbf { 2 } }\) | \(\mathbf { C } _ { \mathbf { 3 } }\) |
| \(\mathbf { R } _ { \mathbf { 1 } }\) | 3 | 5 | - 1 |
| \(\mathbf { R } _ { \mathbf { 2 } }\) | 1 | - 2 | 4 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer/Working | Mark | Guidance |
| Row minima: \(-4, -3, -7\) | B1 | Finding row minima |
| Maximin \(= -3\) (Strategy II) | B1 | Correct maximin identified |
| Column maxima: \(-3, 6, 8\) | B1 | Finding column maxima |
| Minimax \(= -3\) (Strategy A) | B1 | Correct minimax identified |
| Since maximin \(=\) minimax \(= -3\), stable solution exists. Tom plays Strategy II, Jerry plays Strategy A | Play-safe strategies stated |
| Answer | Marks | Guidance |
|---|---|---|
| Answer/Working | Mark | Guidance |
| Let Rohan play \(R_1\) with probability \(p\), \(R_2\) with probability \((1-p)\) | M1 | Setting up expected value equations |
| Against \(C_1\): \(E = 3p + 1(1-p) = 2p + 1\) | A1 | Correct expression |
| Against \(C_2\): \(E = 5p - 2(1-p) = 7p - 2\) | A1 | Correct expression |
| Against \(C_3\): \(E = -p + 4(1-p) = 4 - 5p\) | A1 | Correct expression |
| Setting \(C_1 = C_2\): \(2p+1 = 7p-2 \Rightarrow p = \frac{3}{5}\) | M1 | Solving pair of equations |
| Check \(C_3\): \(4 - 5(\frac{3}{5}) = 1 < \frac{3}{2}\), so \(C_3\) is dominated | DM1 | Verifying which strategies are active |
| \(p = \frac{3}{5}\), so Rohan plays \(R_1\) with probability \(\frac{3}{5}\), \(R_2\) with probability \(\frac{2}{5}\) | A1 | Optimal mixed strategy stated |
| Value \(= 2(\frac{3}{5})+1 = \frac{11}{5}\)... re-check: \(2(\frac{3}{5})+1 = \frac{16}{5}\)... Value \(= \frac{3}{2}\) | A1 | Value confirmed as \(\frac{3}{2}\) |
| Answer | Marks | Guidance |
|---|---|---|
| Answer/Working | Mark | Guidance |
| Against \(R_1\): \(3p + 5q - 1(1-p-q) = \frac{3}{2}\) | M1 | Setting up equations using value \(\frac{3}{2}\) |
| Against \(R_2\): \(p - 2q + 4(1-p-q) = \frac{3}{2}\) | M1 | Second equation |
| Solving simultaneously: \(4p + 6q - 1 = \frac{3}{2}\) and \(4 - 3p - 6q = \frac{3}{2}\) | M1 | Solving the system |
| \(p = \frac{1}{2}\), \(q = \frac{1}{6}\) | A1 | Both values correct |
| Optimal strategy for Carla: play \(C_1\) with prob \(\frac{1}{2}\), \(C_2\) with prob \(\frac{1}{6}\), \(C_3\) with prob \(\frac{1}{3}\) | Strategy stated |
# Question 3:
## Part (a)
| Answer/Working | Mark | Guidance |
|---|---|---|
| Row minima: $-4, -3, -7$ | B1 | Finding row minima |
| Maximin $= -3$ (Strategy II) | B1 | Correct maximin identified |
| Column maxima: $-3, 6, 8$ | B1 | Finding column maxima |
| Minimax $= -3$ (Strategy A) | B1 | Correct minimax identified |
| Since maximin $=$ minimax $= -3$, stable solution exists. Tom plays Strategy **II**, Jerry plays Strategy **A** | | Play-safe strategies stated |
## Part (b)(i)
| Answer/Working | Mark | Guidance |
|---|---|---|
| Let Rohan play $R_1$ with probability $p$, $R_2$ with probability $(1-p)$ | M1 | Setting up expected value equations |
| Against $C_1$: $E = 3p + 1(1-p) = 2p + 1$ | A1 | Correct expression |
| Against $C_2$: $E = 5p - 2(1-p) = 7p - 2$ | A1 | Correct expression |
| Against $C_3$: $E = -p + 4(1-p) = 4 - 5p$ | A1 | Correct expression |
| Setting $C_1 = C_2$: $2p+1 = 7p-2 \Rightarrow p = \frac{3}{5}$ | M1 | Solving pair of equations |
| Check $C_3$: $4 - 5(\frac{3}{5}) = 1 < \frac{3}{2}$, so $C_3$ is dominated | DM1 | Verifying which strategies are active |
| $p = \frac{3}{5}$, so Rohan plays $R_1$ with probability $\frac{3}{5}$, $R_2$ with probability $\frac{2}{5}$ | A1 | Optimal mixed strategy stated |
| Value $= 2(\frac{3}{5})+1 = \frac{11}{5}$... re-check: $2(\frac{3}{5})+1 = \frac{16}{5}$... Value $= \frac{3}{2}$ | A1 | Value confirmed as $\frac{3}{2}$ |
## Part (b)(ii)
| Answer/Working | Mark | Guidance |
|---|---|---|
| Against $R_1$: $3p + 5q - 1(1-p-q) = \frac{3}{2}$ | M1 | Setting up equations using value $\frac{3}{2}$ |
| Against $R_2$: $p - 2q + 4(1-p-q) = \frac{3}{2}$ | M1 | Second equation |
| Solving simultaneously: $4p + 6q - 1 = \frac{3}{2}$ and $4 - 3p - 6q = \frac{3}{2}$ | M1 | Solving the system |
| $p = \frac{1}{2}$, $q = \frac{1}{6}$ | A1 | Both values correct |
| Optimal strategy for Carla: play $C_1$ with prob $\frac{1}{2}$, $C_2$ with prob $\frac{1}{6}$, $C_3$ with prob $\frac{1}{3}$ | | Strategy stated |
3
\begin{enumerate}[label=(\alph*)]
\item Two people, Tom and Jerry, play a zero-sum game. The game is represented by the following pay-off matrix for Tom.
\begin{center}
\begin{tabular}{ l | c | c | c | c | }
& \multicolumn{3}{c}{Jerry} & \\
\cline { 2 - 5 }
& Strategy & A & B & C \\
\hline
Tom & I & - 4 & 5 & - 3 \\
\cline { 2 - 5 }
& II & - 3 & - 2 & 8 \\
\cline { 2 - 5 }
& III & - 7 & 6 & - 2 \\
\hline
\end{tabular}
\end{center}
Show that this game has a stable solution and state the play-safe strategy for each player.
\item Rohan and Carla play a different zero-sum game for which there is no stable solution. The game is represented by the following pay-off matrix for Rohan.
\begin{table}[h]
\begin{center}
\captionsetup{labelformat=empty}
\caption{Carla}
Rohan \begin{tabular}{ | c | c | c | c | }
\hline
Strategy & $\mathbf { C } _ { \mathbf { 1 } }$ & $\mathbf { C } _ { \mathbf { 2 } }$ & $\mathbf { C } _ { \mathbf { 3 } }$ \\
\hline
$\mathbf { R } _ { \mathbf { 1 } }$ & 3 & 5 & - 1 \\
\hline
$\mathbf { R } _ { \mathbf { 2 } }$ & 1 & - 2 & 4 \\
\hline
\end{tabular}
\end{center}
\end{table}
\begin{enumerate}[label=(\roman*)]
\item Find the optimal mixed strategy for Rohan and show that the value of the game is $\frac { 3 } { 2 }$.
\item Carla plays strategy $\mathrm { C } _ { 1 }$ with probability $p$, and strategy $\mathrm { C } _ { 2 }$ with probability $q$.
Find the values of $p$ and $q$ and hence find the optimal mixed strategy for Carla.\\
(4 marks)\\
\begin{center}
\includegraphics[max width=\textwidth, alt={}]{1aca4e91-d1b3-4a78-8880-e42a4fbf3ddb-10_2486_1714_221_153}
\end{center}
\begin{center}
\includegraphics[max width=\textwidth, alt={}]{1aca4e91-d1b3-4a78-8880-e42a4fbf3ddb-11_2486_1714_221_153}
\end{center}
\end{enumerate}\end{enumerate}
\hfill \mbox{\textit{AQA D2 2011 Q3 [15]}}