Descriptive Statistics of “timeSeries” structure data using psych package in R

Question

Sorry for this stupid question, I am new to R. I have some in such formate and saved in CSV:

%Y-%m-%d,st1,st2,st3,st4,st5,st6,st7,st8,st9,st10
2005-09-20,38.75,48.625,48.5,23.667,45.5,48.75,18.75,33.25,43.455,76.042
2005-09-21,39.482,49.3,49,23.9,46.15,50.281,18.975,34.125,44.465,78.232
...

I import it in R

library(fPortfolio)

Data <- readSeries(file = "data.csv", header = TRUE, sep = ",")

I want to have some descriptive statistics

library(psych)
describe(Data)

Error in x[!is.na(x[, i]), i] : 
  invalid or not-yet-implemented 'timeSeries' subsetting

Any suggestion?

Answer 1

you probably want to make it a time series first right?

tS <- dummySeries() #make quick dummy time series
describe(tS)  # fails

but

newtS<-as.ts(tS)
describe(newtS)  #works fine giving:

        var  n mean   sd median trimmed  mad  min  max range skew kurtosis   se
Series 1   1 12 0.49 0.25   0.44    0.48 0.29 0.13 0.89  0.76 0.24    -1.52 0.07
Series 2   2 12 0.45 0.28   0.44    0.45 0.42 0.07 0.83  0.77 0.03    -1.74 0.08