lixzhang

* sum of two variables in SAS ;
data output ;
    set input ;
    z = x + y ;
run ;

lixzhang

# sum of two variables in R
input$z <- input$x + input$y

lixzhang

bank_count <- bank
bank_count$condition <- bank_count$Expense > 40
bank_count <- aggregate(bank_count$condition, by = list(bank_count$CardID), FUN = sum)
names(bank_count) <- c("CardID", "over_spend")
bank_count

lixzhang

bank_sum <- aggregate(bank$Expense,
                      by = list(bank$CardID),
                      FUN = sum)
names(bank_sum) <- c("CardID",
                     "total_expense")

lixzhang

bank_mod <- reshape(bank,
                     v.names = "Expense",
                     idvar = "CardID",
                     timevar = "month",
                     direction = "wide")
names(bank_mod) <- gsub("\\.", "", names(bank_mod))
bank_mod <- bank_mod[, c("CardID", "Expense1", "Expense2", "Expense3")]

lixzhang

CardID <- c(1, 1, 1, 2, 2, 2)
month <- c(3, 2, 1, 3, 2, 1)
Expense <- c(100, 50, 70, 20, 30, 60)
bank <- data.frame(cbind(CardID, month, Expense))

lixzhang

data bank_count (keep = CardID over_spend) ;
    set bank ;
    by CardID ;
    retain over_spend ;
    if first.CardID then over_spend = 0 ;
    if Expense > 40 then over_spend = over_spend + 1 ;
    if last.CardID ;
run ;

lixzhang

data bank_sum (keep = CardID total_expense) ;
    set bank ;
    by CardID ;
    retain total_expense ;
    if first.CardID then total_expense = 0 ;
    total_expense = total_expense + Expense ;
    if last.CardID ;
run ;

lixzhang

data bank_mod (drop = i month Expense) ;
    set bank ;
    by CardID ;
    if _n_ = 1 then do ;
        array Expensex(3) Expense1 - Expense3 ;
    end ;
    retain Expense1 - Expense3 ;
    if first.CardID then do ;
        do i = 1 to 3 ;
            Expensex(i) = . ;

lixzhang

data bank ;
input CardID month Expense ;
datalines ;
1 3 100
1 2 50
1 1 70
2 3 20
2 2 30
2 1 60
run ;