X-11/Y2k Quarterly Seasonal Adjustment (Census Method II) - Advanced Tab

Select the Advanced tab of the X-11/Y2k Quarterly Seasonal Adjustment (Census Method II) dialog box to access the options described here.

Element Name Description
Seasonal adjustment The Seasonal adjustment group box contains two options: Additive and Multiplicative. Select an option to determine which type of seasonal adjustment will be performed (refer to X-11 Census Method II Seasonal Adjustment for details).
Additive Select the Additive option button to perform an additive seasonal adjustment. In short, in the additive model, the seasonal, trend-cycle, and irregular (random) components are assumed to combine in an additive fashion (e.g., the seasonal fluctuation adds $1,000,000 to the sales in the holiday month of December).
Multiplicative Select the Multiplicative option button to perform a multiplicative seasonal adjustment. In the multiplicative model, the seasonal, trend-cycle, and irregular (random) components are thought to combine in a multiplicative fashion (e.g., due to the seasonal component, in December sales increase by 30%, i.e., 1.3 times the baseline).
Dates (start of series) Use the options in the Dates (start of series) group box to determine the start date of the series. The start date not only affects the labeling of the output tables, but can also substantially affect the computations.
From variable Select the From variable option button to display the standard variable selection dialog, in which you select the variable containing the starting date (i.e., the date stored in this variable in the first case of the respective series will be used as the start date of the series).
Variable Click the Variable button to display the standard variable selection dialog, in which you select the variable containing the starting date.
From Quarter/Year Select the From Quarter/Year option button to specify the quarter and year in which the series starts.

Note: Specifying dates. You can enter either four-digit years in the range from 1901 through 2099, or two-digit years in the range from 0 through 99. If you specify a two-digit year less than 20, the respective year will be interpreted as 20xx, where xx stands for the two-digit year (e.g., 11 will be interpreted as 2011); if you specify a two-digit year greater than or equal to 20, then the year will be interpreted as 19xx, where xx stands for the two-digit year (e.g., 55 will be interpreted as 1955).

Summary measures (input is seasonally adjusted) If the Summary measures check box is selected, then STATISTICA assumes that the input is already a seasonally adjusted series. In that case, the X-11 procedure can be used to test for residual seasonal variation by completing the standard steps in parts B and C of the computations (see X-11 Result Tables for details).
Printout tables (detail) Use the Printout tables (detail) group box to determine the number of specific X-11 result tables to be computed. See X-11 Result Tables for a description of the specific tables (spreadsheets) that are reported for the Standard, Long, and Full printout details.
Standard (17 tables) Select the Standard (17 tables) option button to display 17 to 27 tables.
Long (27 tables) Select the Long (27 tables) option button to display 27 to 39 tables.
Full (44 tables) Select the Full (44 tables) option button to display 44 to 59 tables.
Selected tables Select the Selected tables option button (or click the adjacent    button) to display the Select the X-11 tables and charts to be displayed/printed dialog box, in which the identifiers for all available tables and charts are listed. Use this dialog to select the specific tables (spreadsheets) that you want displayed and/or printed. If no tables are selected (i.e., do not select any tables and click OK in this dialog), then only the final estimates of the seasonally adjusted series, the seasonal component, irregular component, and trend-cycle component will be appended to the active work area.
Charts Use the Charts group box to determine the specific graphics results to be displayed. Refer also to X-11 result tables for a description of the charts (line plots) computed by the procedure.
None If you select the None option button, then no plots are produced.
Standard If you select the Standard option button, then charts G 1 (final seasonally adjusted series and final trend-cycle components, D 11 and D 12, respectively) and G 2 [final S-I differences or ratios with the extremes, the final S-I differences (ratios) without extremes, and the final seasonal factors; D 8, D 9, and D 10] are computed.
All Select the All option button to display all charts G 1 through G 4 (see X-11 result tables for details).
Selected Select the Selected option button (or click the adjacent    button) to display the Select the X-11 tables and charts to be displayed/printed dialog box, in which all available output charts (and tables) are listed. Select the charts (line plots) to be computed.
Sigma limits for graduating extreme values One of the strengths of the X-11 method is in the treatment of outliers. When estimating the seasonal and trend-cycle components, irregular values are assigned different weights depending on their deviation from 0 (zero; additive model) or 1 (multiplicative model). See also X-11 result tables (table B 4) for details concerning this procedure.
Full weight The Full weight value determines the multiple of σ (Sigma; the estimate of the trend-cycle standard deviation) below which a value will be assigned a full weight.
Zero weight The Zero weight value determines the multiple of σ above which a zero weight will be assigned. Between these two cut-off values, linearly graduated weights are assigned.
Adjust trend-cycle for strikes (extremes) Select the Adjust trend-cycle for strikes (extremes) check box to incorporate an additional adjustment for strikes or other extreme outliers in the computation of the trend-cycle component (X-11 result tables B 7). This adjustment for extremes may substantially reduce the effect of major prolonged strikes or similar irregular occurrences on the B 7 and subsequent trend-cycle estimates. However, Shiskin, Young, and Musgrave (1967) caution that estimates near sharp business cycle peaks or troughs will also be affected by this adjustment.
Other transformations & plots Click the Other transformations & plots button to display the Transformation of Variables dialog, which contains options to perform a wide variety of transformations on the data. The transformed series will be appended to the active work area.