R 列举具有不同概率的一系列伯努利试验的所有可能组合概率

假设我有一系列n个独立伯努利试验成功的概率，p1到pn，这样p1！=p2！=…！=请注意。为每个试验指定一个唯一的名称

    p <- c(0.5, 0.12, 0.7, 0.8, .02)
    a <- c("A","B","C","D","E")

---

p快速计算的关键是在对数概率空间中进行计算，以便树的每个分支的乘积都是一个可以计算为矩阵乘法的内部和的和。以这种方式，所有分支都可以矢量化方式一起计算

首先，我们构造所有分支的枚举。为此，我们使用
R.utils
包中的
intToBin
函数：

library(R.utils)
enum.branches <- unlist(strsplit(intToBin(seq_len(2^n)-1),split=""))

结果在一个矩阵中，其中每列都是概率树分支的结果

现在，如果
enum.branchs==“1”
和
log（1-p）
，则使用它来构造与
enum.branchs
大小相同的日志概率矩阵，其中值为
log（p）
。对于您的数据，使用
p只需在base R中尝试以下操作：

p <- c(0.5, 0.12, 0.7, 0.8, .02)
a <- c("A","B","C","D","E")
n <- length(p)
apply(expand.grid(replicate(n,list(0:1)))[n:1], 1, 
                  function(x) prod(p[which(x==1)])*prod(1-p[which(x==0)]))

#[1] 0.025872 0.000528 0.103488 0.002112 0.060368 0.001232 0.241472 0.004928 0.003528 0.000072 0.014112 0.000288 0.008232 0.000168 0.032928 0.000672 0.025872
#[18] 0.000528 0.103488 0.002112 0.060368 0.001232 0.241472 0.004928 0.003528 0.000072 0.014112 0.000288 0.008232 0.000168 0.032928 0.000672

p太棒了，这就像一个符咒！作为奖励，我还了解到对数概率之和的指数等于概率的乘积。整洁的仅供参考，在我的macbook air上运行双1.3 GHz：14.255、2.001、17.652。不错！伟大的解决方案，比@aichao nice解决方案更快，因为它消除了“枚举”步骤。
matrix(enum.branches, nrow=n)
##     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13] [,14] [,15] [,16] [,17]
##[1,] "0"  "0"  "0"  "0"  "0"  "0"  "0"  "0"  "0"  "0"   "0"   "0"   "0"   "0"   "0"   "0"   "1"  
##[2,] "0"  "0"  "0"  "0"  "0"  "0"  "0"  "0"  "1"  "1"   "1"   "1"   "1"   "1"   "1"   "1"   "0"  
##[3,] "0"  "0"  "0"  "0"  "1"  "1"  "1"  "1"  "0"  "0"   "0"   "0"   "1"   "1"   "1"   "1"   "0"  
##[4,] "0"  "0"  "1"  "1"  "0"  "0"  "1"  "1"  "0"  "0"   "1"   "1"   "0"   "0"   "1"   "1"   "0"  
##[5,] "0"  "1"  "0"  "1"  "0"  "1"  "0"  "1"  "0"  "1"   "0"   "1"   "0"   "1"   "0"   "1"   "0"  
##     [,18] [,19] [,20] [,21] [,22] [,23] [,24] [,25] [,26] [,27] [,28] [,29] [,30] [,31] [,32]
##[1,] "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"  
##[2,] "0"   "0"   "0"   "0"   "0"   "0"   "0"   "1"   "1"   "1"   "1"   "1"   "1"   "1"   "1"  
##[3,] "0"   "0"   "0"   "1"   "1"   "1"   "1"   "0"   "0"   "0"   "0"   "1"   "1"   "1"   "1"  
##[4,] "0"   "1"   "1"   "0"   "0"   "1"   "1"   "0"   "0"   "1"   "1"   "0"   "0"   "1"   "1"  
##[5,] "1"   "0"   "1"   "0"   "1"   "0"   "1"   "0"   "1"   "0"   "1"   "0"   "1"   "0"   "1"  

logp <- matrix(ifelse(enum.branches == "1", rep(log(p), 2^n), rep(log(1-p), 2^n)), nrow=n)

result <- exp(rep(1,n) %*% logp)
##         [,1]     [,2]     [,3]     [,4]     [,5]     [,6]     [,7]     [,8]     [,9]   [,10]
##[1,] 0.025872 0.000528 0.103488 0.002112 0.060368 0.001232 0.241472 0.004928 0.003528 7.2e-05
        [,11]    [,12]    [,13]    [,14]    [,15]    [,16]    [,17]    [,18]    [,19]    [,20]
##[1,] 0.014112 0.000288 0.008232 0.000168 0.032928 0.000672 0.025872 0.000528 0.103488 0.002112
        [,21]    [,22]    [,23]    [,24]    [,25]   [,26]    [,27]    [,28]    [,29]    [,30]
##[1,] 0.060368 0.001232 0.241472 0.004928 0.003528 7.2e-05 0.014112 0.000288 0.008232 0.000168
        [,31]    [,32]
##[1,] 0.032928 0.000672

enum.prob.product <- function(n, p) {
  enum.branches <- unlist(strsplit(intToBin(seq_len(2^n)-1),split=""))
  exp(rep(1,n) %*% matrix(ifelse(enum.branches == "1", rep(log(p), 2^n), rep(log(1-p), 2^n)), nrow=n))
}

n <- 19
p <- runif(n)
system.time(enum.prob.product(n,p))
##   user  system elapsed 
## 24.064   1.470  26.082 

p <- c(0.5, 0.12, 0.7, 0.8, .02)
a <- c("A","B","C","D","E")
n <- length(p)
apply(expand.grid(replicate(n,list(0:1)))[n:1], 1, 
                  function(x) prod(p[which(x==1)])*prod(1-p[which(x==0)]))

#[1] 0.025872 0.000528 0.103488 0.002112 0.060368 0.001232 0.241472 0.004928 0.003528 0.000072 0.014112 0.000288 0.008232 0.000168 0.032928 0.000672 0.025872
#[18] 0.000528 0.103488 0.002112 0.060368 0.001232 0.241472 0.004928 0.003528 0.000072 0.014112 0.000288 0.008232 0.000168 0.032928 0.000672