Question

R的新功能...这是着名的（可能？）概率模拟的公交车开始为空，并在每个站点接收0,1或2名乘客，概率分别为0.5,0.4和0.2。同样在每个站点，每位乘客下车的概率为0.2

我可以看到，在开始时它会跳过一部分代码作为变量＆＃39;乘客＆＃39;从0开始。但我不明白的是runif（1）代码 - 它在这里做了什么？另外一个问题 - 我如何跟踪向量，我想看看它如何存储信息，如表格（对R来说是新的，所以没有线索）。

（此代码模拟10次停止后空巴士的概率）

nreps <- 10000
nstops <- 10
count <- 0
for (i in 1:nreps){
  passengers <- 0
  for (j in 1:nstops){
    if(passengers > 0)
      for (k in 1:passengers)
        if(runif(1) < 0.2)
          passengers <- passengers - 1
    newpass <- sample(0:2, 1, prob=c(0.5, 0.4, 0.1))
    passengers <- passengers + newpass
  }
  if (passengers == 0) count <- count + 1
}
print(count/nreps)

Answer 1

首先，这是每个步骤的解释：

nreps <- 10000
nstops <- 10
count <- 0

for (i in 1:nreps){  #for 10,000 iterations
  passengers <- 0  #start with zero passengers
  for (j in 1:nstops){  #for the number of stops (10) do...
    if(passengers > 0)  #for number of passengers, when number > 0...
      for (k in 1:passengers)  #for each passenger...
        if(runif(1) < 0.2)  #take 1 random val from a uniform dist, and if less than 0.2...
          passengers <- passengers - 1  #subtract 1 from passengers (i.e. n-1)
        newpass <- sample(0:2, 1, prob=c(0.5, 0.4, 0.1))  #take one random sample from 0,1,2, with probability weights 0.5, 0.4, 0.1, respectively
        passengers <- passengers + newpass  #add existing passengers to number of new passengers
  }
  if (passengers == 0) count <- count + 1  #if the number of passengers still equal zero by now, add 1 to the count
}

print(count/nreps)  #number of times when passenger = 0 divided by number of iterations

现在，如果要跟踪每次迭代的进度，可以先设置空变量，然后根据每次迭代的索引填充这些变量，如下所示：

passengers_before <- c()  #empty vector
passengers_after <- c()  #empty vector

for (i in 1:nreps){   #for 10,000 iterations
  passengers <- 0   #start with zero passengers
  passengers_before[i] <- passengers
  for (j in 1:nstops){   #for the number of stops (10) do...
    if(passengers > 0)   #for number of passengers, when number > 0...
      for (k in 1:passengers)   #for each passenger...
        if(runif(1) < 0.2)   #take 1 random val from a uniform dist, and if less than 0.2...
          passengers <- passengers - 1   #subtract 1 from passengers (i.e. n-1)
        newpass <- sample(0:2, 1, prob=c(0.5, 0.4, 0.1))   #take one random sample from 0,1,2, with probability weights 0.5, 0.4, 0.1, respectively
        passengers <- passengers + newpass   #add existing passengers to number of new passengers
  }
  passengers_after[i] <- passengers
  if (passengers == 0) count <- count + 1   #if the number of passengers still equal zero by now, add 1 to the count
}

passenger_sample <- data.frame(passengers_before,passengers_after)

print(count/nreps)

现在，passenger_sample将是一个data.frame，其行数等于迭代次数（例如10,000），它将有两列：passengers_before和passengers_after。

这是第一批第一行：

head(passenger_sample)

  passengers_before passengers_after
1                 0                2
2                 0                4
3                 0                4
4                 0                4
5                 0                1
6                 0                2

需要帮助理解这个概率代码

1 个答案: