Finding the second highest value in quantmod by time period

Question

I am trying to find the maximal and the second largest value during a 6 months interval. I am using runMax to find the first value but I can't figure out how to do for the second one. Here's my code so far:

    library(quantmod)
    library(TTR)
    getSymbols("GOOGL")
    GOOGL_mo<-to.monthly(GOOGL)#to get monthly data
    GOOGL_mo$Max_6mo<-runMax(GOOGL.High, 6)#add a column with the max value during a 6 months period

I would like to add another column with the second largest value to have something like this:

Date        Open      High     Low        Close     Volume   Max6mo  2ndMax6m0
Oct 2016    802.55    839.00   796.23     809.90    35391730 839.00  819.06
Nov 2016    810.87    816.04   743.59     775.88    48353316 839.00  819.06
Dec 2016    778.55    824.30   753.36     792.45    34356689 839.00  824.00
Jan 2017    800.62    867.00   796.89     820.19    36840255 867.00  839.00
Feb 2017    824.00    853.79   812.05     844.93    26492197 867.00  853.79
Mar 2017    851.38    874.42   824.30     847.80    34553208 874.42  867.00
Apr 2017    848.75    935.90   834.60     924.52    28721553 935.90  874.42
May 2017    924.15    965.90   920.80     942.17    21302485 965.90  935.90

Any idea?

Answer 1

Here's one way to do it:

df <- read.table(header=T, text="
            Open      High     Low        Close     Volume   Max6mo  2ndMax6m0
Aug2016    786.67    813.88   785.04     789.85    28857075 813.88  803.94
Sep2016    791.98    819.06   783.50     804.06    31574466 819.06  813.88
Oct2016    802.55    839.00   796.23     809.90    35391730 839.00  819.06
Nov2016    810.87    816.04   743.59     775.88    48353316 839.00  819.06
Dec2016    778.55    824.30   753.36     792.45    34356689 839.00  824.00
Jan2017    800.62    867.00   796.89     820.19    36840255 867.00  839.00
Feb2017    824.00    853.79   812.05     844.93    26492197 867.00  853.79
Mar2017    851.38    874.42   824.30     847.80    34553208 874.42  867.00
Apr2017    848.75    935.90   834.60     924.52    28721553 935.90  874.42
May2017    924.15    965.90   920.80     942.17    21302485 965.90  935.90")
cbind(
  df, 
  rollapplyr(df$High, 6, function(x) 
             setNames(sort(x, decreasing = T)[1:2], c("a","b")), fill = NA)
)
#           Open   High    Low  Close   Volume Max6mo X2ndMax6m0      a      b
# Aug2016 786.67 813.88 785.04 789.85 28857075 813.88     803.94     NA     NA
# Sep2016 791.98 819.06 783.50 804.06 31574466 819.06     813.88     NA     NA
# Oct2016 802.55 839.00 796.23 809.90 35391730 839.00     819.06     NA     NA
# Nov2016 810.87 816.04 743.59 775.88 48353316 839.00     819.06     NA     NA
# Dec2016 778.55 824.30 753.36 792.45 34356689 839.00     824.00     NA     NA
# Jan2017 800.62 867.00 796.89 820.19 36840255 867.00     839.00 867.00 839.00
# Feb2017 824.00 853.79 812.05 844.93 26492197 867.00     853.79 867.00 853.79
# Mar2017 851.38 874.42 824.30 847.80 34553208 874.42     867.00 874.42 867.00
# Apr2017 848.75 935.90 834.60 924.52 28721553 935.90     874.42 935.90 874.42

Answer 2

library(quantmod)
library(TTR)
getSymbols("GOOG",src="google")
GOOG <- to.monthly(GOOG)

thing <- data.frame(max1 = rep(NA, 5), max2 = rep(NA, 5))
for (x in 1:(nrow(GOOG)-5)) {
  max1 <- max(GOOG[x:(x+5), "GOOG.High"])
  max2 <- max(GOOG$GOOG.High[x:(x+5)][GOOG$GOOG.High[x:(x+5)] != max1])
  thing <- rbind(thing, data.frame(max1, max2))
}
GOOG <- cbind(data.frame(GOOG), thing)