option chain web scraping example for AAPL

scrapeOptionChain.R

#load packages
library(quantmod)
library(rvest)

#get underlying stock info and last trade date
symbol <- "AAPL"
priceInfo <- getQuote(symbol)
lastPrice <- priceInfo$Last
date <- as.Date(priceInfo$`Trade Time`)

#settings for moneyness and time to maturity
moneynessBoundaries <- c(0.85,1.15)
ttmBoundaries <- c(7, 183)

#scrape full site
baseUrl <- paste0("https://finance.yahoo.com/quote/",symbol,"/options")
baseHTML <- read_html(baseUrl)

#get available expiries and convert to time to maturity
expiriesUNIX <- baseHTML %>% html_nodes("option") %>% html_attr("value")
expiries <- as.Date((baseHTML %>% html_nodes("option") %>% html_text()), format = "%b %d, %Y")
timeToMats <- as.numeric(expiries - date)

#select applicable expiries
sel <- timeToMats >= ttmBoundaries[1] & timeToMats <= ttmBoundaries[2]

expiriesUNIX <- expiriesUNIX[sel]
expiries <- expiries[sel]
timeToMats <- timeToMats[sel]

#loop over expiries to get calls and puts
calls <- NULL
puts <- NULL

for(i in 1:length(expiriesUNIX)){
  expiryUrl <- paste0(baseUrl,"?date=",expiriesUNIX[i])
  expiryHTML <- read_html(expiryUrl)
  tmpCalls <- expiryHTML %>% html_nodes(".calls") %>% html_table()
  if(length(tmpCalls) > 0){
    tmpCalls <- tmpCalls[[1]]
    #sometimes column names are in uppercase, sometimes not
    colnames(tmpCalls) <- tolower(colnames(tmpCalls))
    #remove thousand separator and convert to numeric if applicable
    tmpCalls$strike <- as.numeric(gsub(",","",tmpCalls$strike))
    #add time to maturity
    tmpCalls$ttm <- timeToMats[i]
    #calculate moneyness
    tmpCalls$moneyness <- lastPrice/tmpCalls$strike
    #convert yahoo finance IV to numeric
    tmpCalls$ivOrig <- as.numeric(gsub("%","",gsub(",","",tmpCalls$`implied volatility`)))/100
    calls <- rbind(calls, tmpCalls)
  }
  
  tmpPuts <- expiryHTML %>% html_nodes(".puts") %>% html_table()
  if(length(tmpPuts) > 0){
    tmpPuts <- tmpPuts[[1]]
    #sometimes column names are in uppercase, sometimes not
    colnames(tmpPuts) <- tolower(colnames(tmpPuts))
    #remove thousand separator and convert to numeric if applicable
    tmpPuts$strike <- as.numeric(gsub(",","",tmpPuts$strike))
    #add time to maturity
    tmpPuts$ttm <- timeToMats[i]
    #calculate moneyness
    tmpPuts$moneyness <- tmpPuts$strike/lastPrice
    #convert yahoo finance IV to numeric
    tmpPuts$ivOrig <- as.numeric(gsub("%","",gsub(",","",tmpPuts$`implied volatility`)))/100
    puts <- rbind(puts, tmpPuts)
  }
}

#select only calls within the applicable moneyness boundaries
calls <- calls[calls$moneyness >= moneynessBoundaries[1] & calls$moneyness <= moneynessBoundaries[2],]
#select only calls that have traded during the last 5 minutes of the stocks last trading day
calls <- calls[strptime(calls$`last trade date`,format = "%Y-%m-%d %I:%M%p") >= strptime(paste0(date," 3:55PM EDT"), format = "%Y-%m-%d %I:%M%p"),]

#select only puts within the applicable moneyness boundaries
puts <- puts[puts$moneyness >= moneynessBoundaries[1] & puts$moneyness <= moneynessBoundaries[2],]
#select only puts that have traded during the last 5 minutes of the stocks last trading day
puts <- puts[strptime(puts$`last trade date`,format = "%Y-%m-%d %I:%M%p") >= strptime(paste0(date," 3:55PM EDT"), format = "%Y-%m-%d %I:%M%p"),]