- UID
- 223282
- 帖子
- 241
- 主题
- 126
- 注册时间
- 2011-7-11
- 最后登录
- 2013-8-21
|
2#
发表于 2011-7-13 17:04
| 只看该作者
I can't remember the exact form its presented in in the CFAI material, but I prefer to rearrange the equation as:
P(A) * P(B|A) = P(B) * P(A|B)
If you rearrange in this way, Bayes theorem becomes an obvious conclusion. Consider that in this form, either side of the equation gives you the probability of both A and B occurring simultaneously. In order to calculate the probability that both occur simultaneously, you can approach the problem in one of two ways (each "approach" is one side of the equation).
The first approach (left side) focuses on event A. If we are attempting to calculate the probability that both A and B occur, then we start with the probability that A occurs, and hence we have P(A). Given that A has occurred, we then multiply by the probability that B has occurred, given that we know A has occurred, or P(B|A). Multiplying these together results in the probability that both events occur simultaneously.
The second approach (right side) is the same, but starts with event B.
Edited 2 time(s). Last edit at Sunday, July 3, 2011 at 12:08AM by RockGuitar417. |
|