Sometimes missing values should be interpreted as truly unknown values and retained as missing values in the data set. The forecasting models used by the ESM procedure can effectively handle missing values (see the section Missing Value Modeling Issues). However, sometimes missing values are known, such as when missing values are created from accumulation and represent no observed values for the variable. In this case, the value for the period should be interpreted as zero (no values), and the SETMISSING=0 option should be used to cause PROC ESM to recode missing values as zero. In other cases, missing values should be interpreted as global values, such as minimum or maximum values of the accumulated series. The accumulated and missing-value-recoded time series is used in subsequent analyses in PROC ESM.