Simple Regression Analysis was used to develop a four-month, time-based linear regression model of the CPU time consumed by the payroll application. Figure 6-10 provides the control parameters that we used for the study and Figure 6-11 shows the plot that was produced for the specified model.
/------------------------ Simple Regression Analysis -------------------------\ | Command ===> | | | |Composing CA MICS Inquiry: PAYSMR - Simple Regression of Payroll Appl. | | | |Report title ===> Payroll CPU requirements | | | |Selection criteria: | | Input file ===> PAY | | Start date ===> _________ (ddmonyyyy) | | SYSID ===> TSOA | | Zone ===> _________ | | No. weeks ===> ____ (1-9999) | | Specify CAPAPU Values ===> _ (Y/N) | | | |Plot title ===> ________________________________________ | |Forecast data element ===> JOBTCBTM | |Axis label ===> CPU TIME | |Forecast length (weeks) ===> 4 (1-9999) | | | |Plot type ===> PRINTER (COLOR/PRINTER) | |Specify color graphics parms===> N (Y/N/R) | | Color graphics format set ===> PENPLOT | | | \------------------------------------------------------------------------------/ Figure 6-10. Application Tracking Control Parameters
PAYROLL CPU REQUIREMENTS 21:51 WEDNESDAY, OCTOBER 27, 1982 72 SYSID: TSOA ZONE: A PLOT OF JOBTCBTM*DATE LEGEND: A = 1 OBS, B = 2 OBS, ETC. PLOT OF PX*DATE SYMBOL USED IS * PLOT OF EST*DATE SYMBOL USED IS + 21:31:40.00 + A 20:50:00.00 + 20:08:20.00 + A A C P 19:26:40.00 + U * * * T * I * M 18:45:00.00 + + + E + + 18:03:20.00 + 17:21:40.00 + A A 16:40:00.00 + --+-----------+-----------+-----------+-----------+-----------+-----------+-----------+-----------+-- 31MAY97 30JUN97 31JUL97 31AUG97 30SEP97 31OCT97 30NOV97 31DEC97 31JAN98 MONTH ENDING
Figure 6-11. Simple Regression Payroll CPU Time Forecast
|
Copyright © 2014 CA.
All rights reserved.
|
|