multiple time series using ts () function wrong data

Question

i have a monthly time series data containing 528 rows and i use ts()function as follows

oilfulldata <- ts(data$logOil, start = c(1976,01), end = c(2019,12), frequency = 12 )

this function is working proberly and i get the values stored in oilfulldata as i see them in the excel sheet i import the data from

head(Oilfulldata) [1] 1.080266 1.082785 1.085291 1.085291 1.085291 1.085291

second i try to make multiple time series from different dates as follows

Oildata1 <- ts(data$logOil, start = c(1976,01), end = c(1999,12), frequency = 12 )

Oildata2 <- ts(data$logOil, start = c(2002,01), end = c(2019,12), frequency = 12 )

first code also is also working proberly and get the values like i see them in the excel sheet i import the data from

head(Oildata1) [1] 1.080266 1.082785 1.085291 1.085291 1.085291 1.085291

second code is my problem although i get no error but stored data is worng

head(Oildata2) [1] 1.080266 1.082785 1.085291 1.085291 1.085291 1.085291

stored data shows the data from 1976,01 although i specified another start date

can anyone tell me whats going on here ?

Answer 1

You need the window() function:

Oildata1 <- window(oilfulldata, start = c(1976,01), end = c(1999,12))
Oildata2 <- window(oilfulldata, start = c(2002,01), end = c(2019,12))

The ts() function is just grabbing enough observations from the start of the data$logOil vector to match the start, end and frequency arguments. It has no way of knowing what time periods the observations correspond to.