The parameters of this topic tell ProVal how to determine the age, hire age (used, generally, to compute service) and sex of members. The record’s age and sex at each valuation date are then used for table look-ups of decrement probabilities, election probabilities and rates of salary growth, as well as to group the results for reports. You may check the box to Use Census Specifications. If this option is selected, then you will be required to select Census Specifications for each database included under the Databases & Census Specifications topic of the Experience Study. When the Experience Study is run, ProVal will look at the appropriate Census Specifications to determine which fields are used to determine age, service and sex. (For more information about these fields, see the Active Data topic of the Census Specifications command.) Alternatively, instead of referring to Census Specifications, you may directly indicate the parameters to be used to determine Date of birth (or attained age), Date of hire (or hire age), Sex and the codes for Male and Female. These fields are generally the same as in the Census Specifications for Valuations (and Core Projections). Note, however, that the sex field must be of the coded field type: a numeric field containing the percent male cannot be used. The record’s service at each valuation date (determined by the specified date of hire or hire age) is used for table look-ups of election probabilities, rates of salary growth and decrement probabilities, unless a different definition of service has been specified for the study in the Valuation Assumptions referenced under the Assumptions topic. The record’s service at each valuation date (determined by the specified date of hire or hire age) is also used to group the results for reports.

Records with missing values at a valuation date (after applying Data Defaults, if selected under the Database & Census Specifications topic) for any of these data fields will be excluded from the study for the year beginning on the valuation date.

To avoid distortion of results, a record should have consistent entries in the age and sex fields in all databases. If, for example, a birth date has been corrected in the 1/1/1994 database of a study beginning at 1/1/1992, be sure to correct the record in the 1/1/1992 and 1/1/1993 databases also. Although a change in the entry in the hire age field may be valid (e.g., original hire date replaced by new hire date of a rehired employee), consider whether (e.g., to study select and ultimate termination rates, to determine service-based eligibility correctly) you need to replace the entries in the later years’ databases with an adjusted hire date.

Salaries are required only if you are studying salary growth for continuing actives. Although the Salary field may have the same name in all databases (e.g., SALARY), the salary field name need not be the same (e.g., PAY92 may be selected from the 1/1/1992 database, PAY93 may be selected from the 1/1/1993 database, etc.). Indicate whether the salaries contained in the salary fields are for the years that begin on dates above (e.g., if PAY92 in the 1/1/1992 database contains the annual salary rate as of 1/1/1992, assumed to be the amount of salary actually paid in 1992) or end on dates above (e.g., if PAY92 in the 1/1/1992 database contains salary amount paid in 1991). Be mindful of the fact that salaries must be “mapped” to years the same way in all databases: you will not get accurate results if, for example, the 1/1/1992 database’s salary ends on 1/1/1992 but the 1/1/1993 database’s salary begins on 1/1/1993.

You may choose a Headcount weight from the list of database fields in the current Project. Head counts are multiplied by the value in this field, which may contain age, service, salary, the product of salary and service, an actuarial liability, etc. This weight does not affect the salary growth report, because that report does not contain head counts. Note that if you weight by a field that is blank for some members, those members will be excluded from the Experience Study.