Percentile Ranking - R

Question

I need to divide a time series into 20 equally spaced percentiles and create a ranking for that division by classifying implied volatility levels, in this case.

The time series is composed of daily data from 2011 to 2019, and the percentiles will be built in an annual period, so I’m going to look at a value at the moment t and classify it according to 20 percentiles defined from the historical series of 1 year prior to date t and if that value in t is greater than the highest value observed in the previous year, a category 21 would be created to indicate a very high level.

Answer 1

To create the 20 percentiles equally spaced we can use the function seq(). After that we can use the function quantile() to say the values relative to the percentiles we define.

percentis <- seq(0.05, 1, length.out = 20)
percentis
#>  [1] 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70
#> [15] 0.75 0.80 0.85 0.90 0.95 1.00

set.seed(123)
serie_temporal <- rnorm(1000)
quantile(serie_temporal, percentis)
#>           5%          10%          15%          20%          25% 
#> -1.622584310 -1.267328289 -1.049677448 -0.788690220 -0.628324243 
#>          30%          35%          40%          45%          50% 
#> -0.486452454 -0.348884728 -0.214854912 -0.097388721  0.009209639 
#>          55%          60%          65%          70%          75% 
#>  0.127070557  0.242040647  0.373732181  0.513088142  0.664601867 
#>          80%          85%          90%          95%         100% 
#>  0.841412710  1.019321461  1.254751947  1.676133871  3.241039935

^{Created on 2019-05-10 by the reprex package (v0.2.1)}