The EXPAND Procedure

The EXPAND procedure provides time interval conversion and missing value interpolation for time series.

The features of the EXPAND procedure include

PROC EXPAND enables you to control the end point constraints used to fit interpolating cubic spline curves. In addition, you can transform series before and after interpolation (or without interpolation) using any combination of the following:

    shift by constant    square root          lags
    scale by constant    square               leads
    change sign          logistic             differences
    absolute value       inverse logistic     cumulative sum
    reciprocate          upper bounds         moving sum
    logarithm            lower bounds         moving average
    exponential                               reverse series
The EXPAND procedure offers support for a wide range of time series frequencies:
    YEAR           SEMIMONTH        DAY
    SEMIYEAR       TENDAY           HOUR
    QUARTER        WEEK             MINUTE
    MONTH          WEEKDAY          SECOND

The basic interval types can be repeated or shifted to define a great variety of different frequencies, such as fiscal years, biennial periods, work shifts, and so forth.


For further details, see the SAS/ETS® User's Guide