the probability of observing at least one of n events is less than the sum of the probabilities for each event.



这句话是什么意思,它是基于一个什么统计事实那?想了好久都没搞好。
X1,...,Xn是取值0,1的随机变量,那么上面的意思就是:

1-P(X1=0,...,Xn=0)<=P(X1=1)+...+P(Xn=1),

当是iid的情形,就简化成如下事实:

1-(1-p)^n<=np, 0<=p<=1
和事件发生(即至少有一个发生)的概率不超过各事件发生的概率的和。
"the probability of observing at least one of n events is less than the sum of the probabilities for each event."



When the n events are mutually exclusive, the two are equal. Otherwise, the former is less than the later because of the overlaps. It can be shown easily by Venn Diagram.
这个就是概率中的半可加性呀,P(A1∪...∪An)<=∑P(Ai)  i=1,...,n
9 天 后
事件并的概率小于等于事件概率的和 [s:17]
[quote]引用第4楼xjuchenwei于2009-10-22 22:41发表的 :

这个就是概率中的半可加性呀,P(A1∪...∪An)<=∑P(Ai)  i=1,...,n

[/quote]



补充4楼:



且n可取正无穷.



但这个可能不叫半可加性. 而只是概率的可列可加性+非负性的一个结果.



可列可加性说如果A1,A2,...两两互斥, 则上式变成等式.



一般来说半可加性说即使两两互斥, 上式还是可以维持小于等于的形式.



当然A1,A2,...不一定两两互斥, 于是把他们的并写成若干互斥的集合的并时要去掉它们非空的交, 这些非空的交集的概率须非负, 从而有了小于等于号.