User manual SPSS TRENDS 10.0

[. . . ] SPSS Trends 10. 0 TM For more information about SPSS® software products, please visit our WWW site at http://www. spss. com or contact Marketing Department SPSS Inc. 233 South Wacker Drive, 11th Floor Chicago, IL 60606-6307 Tel: (312) 651-3000 Fax: (312) 651-3668 SPSS is a registered trademark and the other product names are the trademarks of SPSS Inc. No material describing such software may be produced or distributed without the written permission of the owners of the trademark and license rights in the software and the copyrights in the published materials. The SOFTWARE and documentation are provided with RESTRICTED RIGHTS. [. . . ] The series step135 takes its step at week 135; step136 takes its step the following week. After week 136, both of these dummy variables contribute to the level of the series. Specifying Predictor Variables in ARIMA The Trends ARIMA procedure allows you to specify one or more predictor variables (also called regressors) for the series you are analyzing. ARIMA treats these predictors much like predictor variables in regression analysis--it estimates the coefficients for them that best fit the data. We will use the same two predictor variables, step135 and step136, for both Crest market share and Colgate market share. We expect positive coefficients for both predictor variables in the Crest model and negative coefficients in the Colgate model. The sum of the Crest coefficients will represent the total increase in Crest market share over the two-week period, and the sum of the Colgate coefficients will represent the total decrease in Colgate market share. Estimating the Models To estimate the intervention analysis model, from the menus choose: Analyze Time Series ARIMA. . . This opens the ARIMA dialog box, as shown in Figure 10. 6. 142 Chapter 10 Figure 10. 6 ARIMA dialog box · Move colgate into the Dependent box. · Move step135 and step136 into the Independent(s) list. · Specify 1 for the Difference parameter d, and specify 1 for the Moving Average parameter q. · Deselect Include constant in model. Since an ARIMA(0, 1, 1) model analyzes differences and neither series showed a long-term trend (aside from the effect of the intervention), we expect the average differences to be 0. To obtain the same analysis for crest, you can simply go to the ARIMA dialog box, move colgate out of the Dependent list, and move crest in. (If you want to reduce processing time for this second analysis, click Options, and in the Initial Values for Estimation group in the ARIMA Options dialog box select Apply from previous model, as explained in Chapter 8. ) Figure 10. 7 and Figure 10. 8 show the results for the intervention analysis. An Effective Decay-Preventive Dentifrice: Intervention Analysis 143 Figure 10. 7 Intervention analysis for Colgate Split group number: 1 Series length: 276 No missing data. Melard's algorithm will be used for estimation. Termination criteria: Parameter epsilon: . 001 Maximum Marquardt constant: 1. 00E+09 SSQ Percentage: . 001 Maximum number of iterations: 10 Initial values: MA1 STEP135 STEP136 . 57105 -. 06619 -. 06061 Marquardt constant = . 001 Adjusted sum of squares = . 63460269 Iteration History: Iteration 1 2 3 Adj. Sum of Squares . 59567279 . 59430370 . 59427067 Marquardt Constant . 00100000 . 00010000 . 00001000 Conclusion of estimation phase. Estimation terminated at iteration number 4 because: Sum of squares decreased by less than . 001 percent. FINAL PARAMETERS: Number of residuals Standard error Log likelihood AIC SBC 275 . 04665299 453. 65255 -901. 3051 -890. 45479 Analysis of Variance: Residuals DF 272 Adj. Sum of Squares . 59426928 Residual Variance . 00217650 Variables in the Model: MA1 STEP135 STEP136 B . 80588760 -. 05245968 -. 06085701 SEB . 03671173 . 04665299 . 04665299 T-RATIO 21. 951775 -1. 124466 -1. 304461 APPROX. . 00000000 . 26180685 . 19317882 The following new variables are being created: Name FIT_1 ERR_1 LCL_1 UCL_1 SEP_1 Label Fit for COLGATE from ARIMA, MOD_8 NOCON Error for COLGATE from ARIMA, MOD_8 NOCON 95% LCL for COLGATE from ARIMA, MOD_8 NOCON 95% UCL for COLGATE from ARIMA, MOD_8 NOCON SE of fit for COLGATE from ARIMA, MOD_8 NOCON 144 Chapter 10 Figure 10. 8 Intervention analysis for Crest Split group number: 1 Series length: 276 No missing data. Melard's algorithm will be used for estimation. Termination criteria: Parameter epsilon: . 001 Maximum Marquardt constant: 1. 00E+09 SSQ Percentage: . 001 Maximum number of iterations: 10 Initial values: MA1 STEP135 STEP136 . 63926 . 06103 . 10316 Marquardt constant = . 001 Adjusted sum of squares = . 53437524 Iteration History: Iteration 1 2 Adj. Sum of Squares . 51922627 . 51921252 Marquardt Constant . 00100000 . 00010000 Conclusion of estimation phase. [. . . ] Always check the ACF and PACF of the residuals, in case your identification is wrong. Bear in mind that: · Seasonal processes show these patterns at the seasonal lags (the multiples of the seasonal period). · You are entitled to treat nonsignificant values as zero. That is, you can ignore values that lie within the confidence intervals on the plots. [. . . ]