Model Comparison ---------------- Data variable: AUTOADJ/CPI Number of observations = 314 Start index = 1/70 Sampling interval = 1.0 month(s) Length of seasonality = 12 Number of periods withheld for validation: 26 Model Comparison ---------------- Data variable: AUTOADJ/CPI Number of observations = 314 Start index = 1/70 Sampling interval = 1.0 month(s) Length of seasonality = 12 Number of periods withheld for validation: 26 Models ------ (A) Constant mean = 1.40164 + 1 regressor (B) Random walk (C) ARIMA(0,1,0) with constant (D) Simple exponential smoothing with alpha = 0.4753 (E) Brown's linear exp. smoothing with alpha = 0.2095 Estimation Period Model MSE MAE MAPE ME MPE ------------------------------------------------------------------------ (A) 4.10166 1.64644 8.3251 -1.43836E-14 -1.041 (B) 1.88523 0.983196 4.77414 0.0503052 0.0388061 (C) 1.8893 0.980477 4.76469 -0.00409051 -0.232459 (D) 1.51792 0.91963 4.50505 0.0997052 0.221245 (E) 1.57124 0.947813 4.69309 0.0256061 -0.0466592 Model RMSE RUNS RUNM AUTO MEAN VAR ----------------------------------------------- (A) 2.02526 ** *** *** OK OK (B) 1.37304 OK *** *** OK *** (C) 1.37452 OK *** *** OK *** (D) 1.23204 OK OK ** OK *** (E) 1.25349 ** ** ** OK ** Validation Period Model MSE MAE MAPE ME MPE ------------------------------------------------------------------------ (A) 10.6607 3.05438 9.9606 3.05438 9.9606 (B) 1.51437 1.12022 3.7068 0.198035 0.542558 (C) 1.49579 1.11603 3.69692 0.14364 0.36204 (D) 1.1757 0.90882 2.97726 0.389334 1.18069 (E) 1.06318 0.840403 2.78052 -0.0247549 -0.205443 Key: RMSE = Root Mean Squared Error RUNS = Test for excessive runs up and down RUNM = Test for excessive runs above and below median AUTO = Box-Pierce test for excessive autocorrelation MEAN = Test for difference in mean 1st half to 2nd half VAR = Test for difference in variance 1st half to 2nd half OK = not significant (p >= 0.10) * = marginally significant (0.05 < p <= 0.10) ** = significant (0.01 < p <= 0.05) *** = highly significant (p <= 0.01) The StatAdvisor --------------- This table compares the results of five different forecasting models. You can change any of the models by pressing the alternate mouse button and selecting Analysis Options. Looking at the error statistics, the model with the smallest mean squared error (MSE) during the estimation period is model D. The model with the smallest mean absolute error (MAE) is model D. The model with the smallest mean absolute percentage error (MAPE) is model D. During the validation period, the model with the smallest mean squared error (MSE) is model E. The model with the smallest mean absolute error (MAE) is model E. The model with the smallest mean absolute percentage error (MAPE) is model E. You can use these results to select the most appropriate model for your needs. The table also summarizes the results of five tests run on the residuals to determine whether each model is adequate for the data. An OK means that the model passes the test. One * means that it fails at the 90% confidence level. Two *'s means that it fails at the 95% confidence level. Three *'s means that it fails at the 99% confidence level. Note that the currently selected model, model C, passes 2 tests. Since one or more tests are statistically significant at the 95% or higher confidence level, you should seriously consider selecting another model. Key: RMSE = Root Mean Squared Error RUNS = Test for excessive runs up and down RUNM = Test for excessive runs above and below median AUTO = Box-Pierce test for excessive autocorrelation MEAN = Test for difference in mean 1st half to 2nd half VAR = Test for difference in variance 1st half to 2nd half OK = not significant (p >= 0.10) * = marginally significant (0.05 < p <= 0.10) ** = significant (0.01 < p <= 0.05) *** = highly significant (p <= 0.01) The StatAdvisor --------------- This table compares the results of five different forecasting models. You can change any of the models by pressing the alternate mouse button and selecting Analysis Options. Looking at the error statistics, the model with the smallest mean squared error (MSE) during the estimation period is model C. The model with the smallest mean absolute error (MAE) is model C. The model with the smallest mean absolute percentage error (MAPE) is model C. During the validation period, the model with the smallest mean squared error (MSE) is model D. The model with the smallest mean absolute error (MAE) is model D. The model with the smallest mean absolute percentage error (MAPE) is model D. You can use these results to select the most appropriate model for your needs. The table also summarizes the results of five tests run on the residuals to determine whether each model is adequate for the data. An OK means that the model passes the test. One * means that it fails at the 90% confidence level. Two *'s means that it fails at the 95% confidence level. Three *'s means that it fails at the 99% confidence level. Note that the currently selected model, model A, passes 2 tests. Since one or more tests are statistically significant at the 95% or higher confidence level, you should seriously consider selecting another model.