这是一个关于骰子游戏的策略分析问题。你可以最多掷两次六面骰子。首次掷出数字x后，可以选择获得x美元或继续掷骰，但继续后会放弃第一次的点数。如果进行第二次掷骰，只有当点数与第二次相同才能得到x美元。通过期望值计算，发现当停止条件为掷出3时，游戏价值最高，为4.25美元。
Question
You can roll a 6-sided dice up to 2 times. After the first roll, if you get a number x, you can decide to either to get x dollars or to choose to continue rolling. But once you decide to continue, you forgo the number you just rolled. If you get to the second roll, you’ll just get x dollars if the second number is x and the game stops. What is the game worth and what is your strategy?
Answer
官方解答：
用到了期望平滑公式
我的解答：
记第一次点数为
X
1
X_1
X1，第二次点数为
X
2
X_2
X2（若没掷则为0），最终点数为
X
X
X。设策略为
X
1
>
a
X_1>a
X1>a时stop，否则continue。
P
(
X
=
i
)
=
P
(
X
=
i
,
X
1
>
a
)
+
P
(
X
=
i
,
X
1
≤
a
)
=
P
(
X
1
=
i
∣
X
1
>
a
)
P
(
X
1
>
a
)
+
P
(
X
2
=
i
∣
X
1
≤
a
)
P
(
X
1
≤
a
)
=
1
6
−
a
×
I
(
i
>
a
)
×
6
−
a
6
+
1
6
×
a
6
=
1
6
×
I
(
i
>
a
)
+
1
6
×
a
6
\begin{aligned} P(X=i) &= P(X=i,X_1>a)+P(X=i,X_1 \leq a) \\ &= P(X_1=i|X_1>a)P(X_1>a)+P(X_2=i|X_1 \leq a)P(X_1 \leq a) \\ &= \frac{1}{6-a} \times I(i>a) \times \frac{6-a}{6} + \frac{1}{6} \times \frac{a}{6} \\ &= \frac{1}{6} \times I(i>a) + \frac{1}{6} \times \frac{a}{6} \end{aligned}
P(X=i)=P(X=i,X1>a)+P(X=i,X1≤a)=P(X1=i∣X1>a)P(X1>a)+P(X2=i∣X1≤a)P(X1≤a)=6−a1×I(i>a)×66−a+61×6a=61×I(i>a)+61×6a
E
(
X
)
=
∑
i
=
1
a
i
P
(
X
=
i
)
+
∑
i
=
a
+
1
6
i
P
(
X
=
i
)
=
∑
i
=
1
a
a
36
i
+
∑
i
=
a
+
1
6
(
1
6
+
a
36
)
i
=
−
a
2
+
6
a
+
42
12
\begin{aligned} E(X) &= \sum\limits_{i=1}^{a} iP(X=i) + \sum\limits_{i=a+1}^6iP(X=i) \\ &=\sum\limits_{i=1}^a \frac{a}{36}i+\sum\limits_{i=a+1}^6(\frac{1}{6}+\frac{a}{36})i \\ &= \frac{-a^2+6a+42}{12} \end{aligned}
E(X)=i=1∑aiP(X=i)+i=a+1∑6iP(X=i)=i=1∑a36ai+i=a+1∑6(61+36a)i=12−a2+6a+42
当a=3时，
E
(
X
)
E(X)
E(X)最大，为4.25
1981
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
