How to display values in time format in SAS?

Applying Core Functions for Time Formatting

Consider a practical scenario where you have a variable within your data set called duration that holds a time value corresponding to 7:30:00. Although this value is stored internally as seconds (27,000), SAS provides several functions—most notably the PUT function—to manipulate and format this time value into various display formats required for reporting or further analysis. The PUT function is especially versatile as it converts a numeric value (like the duration in seconds) into a character string using a specified format.

Below, we explore the primary SAS functions available for manipulating and formatting time values, showcasing the results based on our example duration of 7:30:00. These functions demonstrate not only how to display the full time string but also how to extract specific components like hours or minutes.

• PUT(duration, time8.) – This function formats the duration value using the TIME8. format. This structure is ideal for standard display, ensuring the output includes hours, minutes, and seconds, with a total field length of 8 characters (including colons).
  • This operation will produce the output: 7:30:00.
• PUT(duration, hhmm.) – This is a specialized time format that suppresses the display of seconds, focusing solely on hours and minutes. It is often preferred in reports where second-level precision is unnecessary.
  • This operation will produce the output: 7:30.
• PUT(duration, hours5.2) – This specific format converts the duration into decimal hours. The 5.2 component indicates a total width of 5 characters, with 2 digits reserved for the decimal component. This is highly useful for calculations or presentations requiring fractional hour representations.
  • This operation will produce the output: 7.50. (Since 30 minutes is 0.5 hours).
• hour(duration) – Unlike the PUT function which formats the entire value, the HOUR function extracts and returns only the integer portion representing the hour component of the time value.
  • This operation will produce the output: 7
• minute(duration) – Similarly, the MINUTE function isolates and returns the integer portion representing the minute component of the time value.
  • This operation will produce the output: 30.
• second(duration) – This function extracts and returns the integer portion representing the second component of the time value.
  • This operation will produce the output: 0.

The following comprehensive example illustrates how to apply these distinct formatting techniques within a SAS data step to produce a derived data set containing all these different time representations side-by-side.

Example: Display Values in Time Formats in SAS

Practical Application: Setting Up the Dataset

To demonstrate the application of these formats, we will begin by creating a sample data set. This data set, named my_data, contains information about the duration it took various athletes to complete a specific task. We use the TIME8. informat within the input statement to ensure the textual time inputs are correctly converted and stored internally as numeric seconds since midnight.

Note the structure of the datalines statement: the durations are entered in the standard HH:MM:SS format, and the time8. informat handles the conversion. This initial step is critical for setting up the time data correctly before any output formatting can occur.

/*create dataset*/
data my_data;
    input athlete $ duration time8.;
    datalines;
A 04:15:00
B 10:09:15
C 7:30:00
D 18:55:00
E 14:23:59
F 23:45:10
;
run;

/*view dataset*/
proc print data=my_data;

When the initial data set my_data is viewed using PROC PRINT without any explicit formatting applied to the duration variable, SAS defaults to displaying the raw internal numeric value. This clearly illustrates how time is stored: as a large integer representing the total seconds since midnight.

The output above confirms that the values are stored as seconds. For instance, the first observation (Athlete A, 04:15:00) is equivalent to 15,300 seconds (4 hours * 3600 seconds/hour + 15 minutes * 60 seconds/minute). This internal representation is necessary for accurate calculations. It is worth reiterating a key fact for time calculations: there are precisely 86,400 seconds in a standard 24-hour day.

Transforming Numeric Time into Diverse Formats

The primary goal is now to take these raw numeric duration values and transform them into various easily digestible time formats suitable for analysis and reporting. We achieve this by creating a new data set, new_data, where we use the PUT function (for format conversion to a character string) and the specific time extraction functions (HOUR, MINUTE, SECOND) to generate seven new columns based on the original duration variable.

Each new variable showcases a different technique for presenting the underlying time data. Variables prefixed with duration_ followed by the format name (e.g., duration_time8) are created using the PUT function along with the relevant format specification (TIME8., HHMM., HOUR5.2). The extraction functions (HOUR, MINUTE, SECOND) directly yield numeric values representing those specific components.

/*create new dataset with duration printed in various time formats*/
data new_data;
    set my_data;
    duration_time8 = put(duration, time8.);
    duration_hhmm = put(duration, hhmm.);
    duration_hour52 = put(duration, hour5.2);
    duration_hour = hour(duration);
    duration_minute = minute(duration);
    duration_second = second(duration);
run;

/*view new dataset*/
proc print data=new_data;

This code block efficiently leverages SAS functions to generate a comprehensive output detailing every facet of time formatting discussed. Running PROC PRINT on new_data allows us to compare the resulting column formats directly.

Interpreting the Formatted Output Columns

Upon execution of the data step and the print procedure, we obtain a new table where the original numeric duration is complemented by six new variables, each representing a distinct formatted version of the time data. Analyzing this output is essential for understanding the utility of each format type.

Reviewing the resulting table, observe how each new column manipulates the presentation of the time value:

• duration_time8: This column applies the standard TIME8. format. It consistently displays the duration in the full HH:MM:SS structure, which is the most comprehensive clock display format.
• duration_hhmm: Utilizing the HHMM. format, this column truncates the seconds display, presenting only hours and minutes (HH:MM). This simplification is highly beneficial for high-level reporting where seconds are noise.
• duration_hour52: This column displays the time as decimal hours (e.g., 7.50 for 7 hours and 30 minutes). This is a character representation derived using the HOURd.f format, making time directly comparable to other fractional numeric data.
• duration_hour: This variable, created by the HOUR() function, contains only the integer value representing the hour component (0-23).
• duration_minute: This variable, created by the MINUTE() function, contains only the integer value representing the minute component (0-59).
• duration_second: This variable, created by the SECOND() function, contains only the integer value representing the seconds component (0-59).

These distinct columns illustrate the dual capability of SAS: using formats via the PUT function for aesthetic display (creating character variables) and using extraction functions (HOUR, MINUTE, SECOND) for isolating specific numeric value components for further mathematical analysis or grouping.

Selecting the Appropriate Time Display Format

The choice of which time format to employ is entirely dependent on the analytical or reporting objective. If the goal is high-precision tracking, the TIME8. format is indispensable. If the analysis requires combining time units with standard numerical calculations—such as averaging productivity per fractional hour—the HOURd.f format (generating duration_hour52) is the most efficient.

Conversely, if the requirement is merely to categorize observations based on the hour of the day or to check if a specific event occurred within a certain minute, the extraction functions (HOUR, MINUTE, SECOND) provide the fastest way to obtain the necessary numeric components without having to manipulate a full time string.

By mastering the application of the TIMEw. format family and the corresponding extraction functions, data professionals can ensure that temporal data in SAS data sets is always presented clearly, accurately, and in the most suitable format for the intended audience or calculation. Feel free to incorporate whichever format best addresses your specific data visualization and analytical needs.

The following tutorials explain how to perform other common tasks in SAS: