When I use strptime function in a dataframe the format of the date looks correct. However, when I then convert to ts (time series) so I can use functions like forecast, the data column looks like gibberish and often hard to see if it is correct or not. Consider the following code....
qtr <- c("03/01/2000","06/01/2000","09/01/2000","12/01/2000","03/01/2001","06/01/2001","09/01/2001","12/01/2001")
qtr <- strptime(as.character(qtr), format = "%m/%d/%Y",tz="EST")
dem <- c(1342,1382,1296, 1330, 1360, 1441,1343,1426)
# create data frame
data <- data.frame(qtr,dem)
data
The output from when I type data gives the following:
# qtr dem
# 1 2000-03-01 1342
# 2 2000-06-01 1382
# 3 2000-09-01 1296
# 4 2000-12-01 1330
As expected. However, when I then convert to a time series as below, i get an extra column.
ts=ts(data, start=c(2000,1), end= c(2001,4),frequency=4)
ts
output:
# qtr dem
# 2000 Q1 951886800 1342
# 2000 Q2 959835600 1382
# 2000 Q3 967784400 1296
# 2000 Q4 975646800 1330
Also, if I do a view (i.e., View(ts)) I can see only the second and 3rd columns and its difficult to figure out what the dates mean. What is the format of the middle column? What do the digits in say 975646800 represent?
If you specify data$dem in the ts-object the result should be as you expect.
qtr <- c("03/01/2000","06/01/2000","09/01/2000","12/01/2000","03/01/2001","06/01/2001","09/01/2001","12/01/2001")
qtr <- strptime(as.character(qtr), format = "%m/%d/%Y",tz="EST")
dem <- c(1342,1382,1296, 1330, 1360, 1441,1343,1426)
# create data frame
data <- data.frame(qtr,dem)
myts <- ts(data$dem, start=c(2000, 3), end=c(2001, 12), frequency=4)
myts
#> Qtr1 Qtr2 Qtr3 Qtr4
#> 2000 1342 1382
#> 2001 1296 1330 1360 1441
#> 2002 1343 1426 1342 1382
#> 2003 1296 1330 1360 1441
Created on 2020-01-29 by the reprex package (v0.3.0)