Converting numeric to date issues

问题

I have a date column that is numeric as follows

I am trying to convert this to real date using this function

 as.Date(df$Date1, origin = "1970-01-01")

However I am seeing that these numbers are converted to these dates which are incorrect.

    Date1_Converted
    1980-12-24
    1984-03-26
    1985-01-16
    1988-06-25
    1989-06-16
    1970-02-09
    1970-02-09
    1970-06-22
    1974-09-01
    1981-05-15

The correct transformation should have been

    Date1_CrctTrnsf
    2005.10.31
    2009.02.11
    2009.12.04        
    2013.05.15
    2014.05.06
    1994.02.22
    1994.02.22 
    1994.08.03
    1999.05.03
    2006.03.22

I am assuming this an origin issue, not sure how to fix this, any help on how to fix this issue is much appreciated.

回答1:

# First I copied the data from your question
df <- read.table(con <-file("clipboard"), header = T)
df

# Convert the format
df1 <- as.Date(df$Date1, origin = "1994-11-08")
df1
> as.data.frame(gsub("-", ".", df1))
   gsub("-", ".", df1)
1           2005.10.31
2           2009.01.11
3           2009.11.23
4           2013.05.02
5           2014.04.23
6           1994.12.17
7           1994.12.17
8           1995.04.29
9           1999.07.09
10          2006.03.22

Note, however, that there seems to be an error in the data (or at least the results you are expecting to get from the data -- how did you produce these numbers?):

The input for row 6 is 39
The input for row 7 is 39 and
The input for row 8 is 172

172 - 39 = 133

However you expect the dates for row 6 and 7 to be 1994.02.22 and the result for row 8 to be 1994.08.03. This is not mathematically possible, as there are 162 days between those 2 dates.

来源：https://stackoverflow.com/questions/32144343/converting-numeric-to-date-issues

标签

date

lubridate