Return element from vector A or B based on value of Vector C or D

Question

I am trying to solve the following problem in R. I have a data frame with 4 columns: currency_1_amt, currency_1_denom, currency_2_amt, currency_2_denom. Here are a few rows from this table

|----------------|------------------|----------------|------------------|
| currency_1_amt | currency_1_denom | currency_2_amt | currency_2_denom |
|----------------|------------------|----------------|------------------|
| 100            | USD              | 620            | CNY              |
| 500            | CNY              | 80.55          | USD              |
| 80.55          | INR              | 1              | USD              |
| 100            | INR              | 9.67           | CNY              |

I am trying to compute a notional value for each row:

• If currency_1_denom == "USD" then notional value=currency_1_amt.

• If currency_2_denom == "USD" then notional value=currency_2_amt.

• If neither currency_1_denom == "USD" nor currency_1_denom == "USD" then notional_value = currency_1_amt * exchange_rate_of_currency_1_to_USD (I have another column in the data frame with the relevant exchange rate.

I am not sure how to do this in R without looping through each row. Heer is some pseudo R code that I thought up for doing this

result = numeric(length(df))
for(j in 1:length(df)) {
    if(df[j,"currency_1_denom"] == "USD")
        result[j] = currency_1_amt
   else if(df[j,"currency_2_denom"] == "USD")
       result[j] = currency_2_amt
   else
       result[j] = currency_1_amt * lookup_exchange_rate(currency_1_denom)

Is there a better (e.g. vectorized) way of accomplishing my task?

Answer 1

Recreating your data:

df<- read.table(text ="currency_1_amt currency_1_denom currency_2_amt  currency_2_denom
100            USD              620             CNY            
500            CNY              80.55           USD            
80.55          INR              1               USD            
100            INR              9.67            CNY",
           header = TRUE, stringsAsFactors = FALSE)

Creating a sample exchange rate just to run the example:

df$exchange_rate_of_currency_1_to_USD <- 1:4

Using ifelse:

df$notional_value <- with(df, ifelse(currency_1_denom == "USD", currency_1_amt, 
                                     ifelse(currency_2_denom == "USD", currency_2_amt,
                                            currency_1_amt * exchange_rate_of_currency_1_to_USD)))

Answer 2

I would do it simply this way:

second <- df$currency_2_denom == "USD"
df$national_value[second] <- df$currency_2_amt[second]
first <- df$currency_1_denom == "USD"
df$national_value[first] <- df$currency_1_amt[first]
third <- (df$currency_1_denom != "USD") & (df$currency_2_denom != "USD")
df$national_value[third] <- df$currency_1_amt[third]*lookup_exchange_rate(currency_1_denom[third])

I change the order on purpose (second condition before first), so if the second condition is met, the entries fulfilling also the first condition would not be overwrite.