Problem with mutating in dplyr a big function with nested functions using the case

I have a function that computes the binomial tree of call or put option found here.I made them all one one function that contain all the other functions described in the link such as :


Binomial_Call = function(S0, K, r, Ti, sigma,N) {
  delta_t=Ti/N
  
  q_prob = function(r, delta_t, sigma) {
    u = exp(sigma*sqrt(delta_t))
    d = exp(-sigma*sqrt(delta_t))
    return((exp(r*delta_t) - d)/(u-d))
  }
  q = q_prob(r=r, delta_t=delta_t, sigma=sigma)
  build_stock_tree = function(S0, sigma, delta_t, N) {
    tree = matrix(0, nrow=N 1, ncol=N 1)
    U = exp(sigma*sqrt(delta_t))
    D = exp(-sigma*sqrt(delta_t))
    for (i in 1:(N 1)) {
      for (j in 1:i) {
        tree[i, j] = S0 * U^(j-1) * D^((i-1)-(j-1))
      }  }
    return(tree)
  }
  tree = build_stock_tree(S0=S0, sigma=sigma, delta_t=delta_t, N=N)
  value_binomial_option = function(tree, sigma, delta_t, r, K) {
    q_prob = function(r, delta_t, sigma) {
      u = exp(sigma*sqrt(delta_t))
      d = exp(-sigma*sqrt(delta_t))
      return((exp(r*delta_t) - d)/(u-d))
    }
    q = q_prob(r, delta_t, sigma)
    option_tree = matrix(0, nrow=nrow(tree), ncol=ncol(tree))
    option_tree[nrow(option_tree),] = pmax(tree[nrow(tree),] - K, 0)
    for (i in (nrow(tree)-1):1) {
      for(j in 1:i) {
        option_tree[i,j]=((1-q)*option_tree[i 1,j]   q*option_tree[i 1,j 1])/exp(r*delta_t)
      }
    }
    return(option_tree)
  }
  option = value_binomial_option(tree, sigma=sigma, delta_t=Ti/N, r=r, K=K)
  #return(list(q=q, stock=tree, option=option, price=option[1,1]))
  price=option[1,1]
  print(price)
}

Now I have a data frame :

stock   = rep(42,4)
strike  = rep(40,4)
risk    = rep(0.1)
time    = rep(0.25)
sigma   = rep(0.2)
N       = rep(60,4)
option  = c(rep(c("C","P"),2)) 
Asset   = c(rep("FTSE",2),"FACEBOOK","TWITTER")
data = tibble(stock,strike,risk,time,sigma,N,option,Asset);data
 A tibble: 4 × 8
  stock strike  risk  time sigma     N option Asset   
  <dbl>  <dbl> <dbl> <dbl> <dbl> <dbl> <chr>  <chr>   
1    42     40   0.1  0.25   0.2    60 C      FTSE    
2    42     40   0.1  0.25   0.2    60 P      FTSE    
3    42     40   0.1  0.25   0.2    60 C      FACEBOOK
4    42     40   0.1  0.25   0.2    60 P      TWITTER

When I try to implement the function in mutate:

data%>%
  mutate(price = case_when(Asset != "FTSE" & option == "C" ~ Binomial_Call(stock,strike,risk,time,sigma,N)))

I receive the following error:

Error in mutate():

! Problem while computing `price = case_when(...)`.
Caused by error in `tree[i, j] <- S0 * U^(j - 1) * D^((i - 1) - (j - 1))`:
! number of items to replace is not a multiple of replacement length
Run `rlang::last_error()` to see where the error occurred.
Warning message:
Problem while computing `price = case_when(...)`.
ℹ numerical expression has 4 elements: only the first used

why ? Why Is that happening ?

CodePudding user response：

The function seems to be not vectorized. So we may use rowwise

library(dplyr)
data %>%
   rowwise %>%
    mutate(price = case_when(Asset != "FTSE" & option == "C" ~
       Binomial_Call(stock,strike,risk,time,sigma,N))) %>% 
   ungroup
# A tibble: 4 × 9
  stock strike  risk  time sigma     N option Asset    price
  <dbl>  <dbl> <dbl> <dbl> <dbl> <dbl> <chr>  <chr>    <dbl>
1    42     40   0.1  0.25   0.2    60 C      FTSE     NA   
2    42     40   0.1  0.25   0.2    60 P      FTSE     NA   
3    42     40   0.1  0.25   0.2    60 C      FACEBOOK  3.53
4    42     40   0.1  0.25   0.2    60 P      TWITTER  NA