DateTime Issue

[sdang]

Hi,

I have issues with these datetime variables.  They contain extra micro seconds.  Because of that I cannot do sum or count based on month/year or use TimeStamp prompt.

Please see the image below for what I meant.  See the last 3 numbers after the seconds.  I want to get rid of those last 4 numbers.

Thanks in advance.

[Jagadishkatam]

Please try to use the informat datetime22. to read and display the datetime as expected

data have;

input x : datetime22.;

format x datetime20.;

cards;

02dec2014:18:05:32.037

;

run;

else, simply try to use the format datetime20. to get the expected display.

Thanks,

Jag

[Tom]

Datetime values are the number of seconds since 1/1/1960.  If you do not want the fractional parts of a second then use ROUND() or INT() function to convert the value to an integer.

[PGStats]

I'm always trying to manipulate date and datetime data in ways that are independent of internal representation. I'm not certain it can always be done but in this case you could use a time constant representing a single second (whatever its value for SAS) in the ROUND function :

data have;

input x : datetime22.;

y = round(x, '00:00:01't);

format x y datetime22.3;

datalines;

02dec2014:18:05:32.037

;

proc print; run;

PG

[Kurt_Bremser]

Like that, as it will work even when SAS changes the datetime representation to nanoseconds since the Big Bang.

[ballardw · Accepted solution]

No manipulation of the value needed for many types of summaries. Assign an appropriated format, such as DTMONYY;

Proc means data=have;

     class IssueDate;

     var < a list of variables to summarize>;

     format IssueDate ddmonyy7.;

run;

The formatted values will be used by most procedures to group values.

[sdang]

Thank you solved my problem.