### Table 3 Percentage of transmitted packets for DPS run with different time series forecasting methods

in sensor

"... In PAGE 7: ...1. Gains in update rate Table3 reports the percentage of packets sent when running the DPS with the CM and AR models with orders from 1 to 5 (AR1; .... In PAGE 8: ...Table3 contains the model that yielded the lowest update rate, and that was consequently selected by the AMS procedure. 5.... ..."

### Table 13 Time series forecast model evaluation, mean relative absolute error Area Model Mean absolute relative error [GWh/day], forecast days ahead average avg. error

in Preface

"... In PAGE 51: ...The model improvement figures in terms of mean absolute error against forecast time horizon, days ahead, are in Table 12 and illustrated in Figure 23. Similar figures of mean relative absolute errors are located in Table13 . Improvements of each version compared to previous are calculated from average daily errors.... ..."

### Table 3: Table showing some of the data sets from real processes used in experiments with forecast of chaotic time series.

1994

"... In PAGE 10: ... Concerning the number of parameters against prediction error, it is clear that standard and K-means RBF requires lot of parameters compared to the other methods. Since much of the discussions in this section has circled around deterministic, simulated systems, we conclude this section with Table3 which summarises some of the real data sets analysed. We have not been able to nd a reasonable way of comparing these data sets, so the reader is referred to the original work.... ..."

Cited by 9

### Table 2: Forecast Accuracy of IPI versus P/B. Forecast Accuracy of Time Series Model: P/B IPI

### Table 1 Performances of the di erent forecasting methods on the test set (time series MG17, h=85)

1998

"... In PAGE 9: ....1.3. Results Table1 reports the results of our system GTS-DEC compared with 5 other forecasting methods. INC is an incremental method for fuzzy models adjustment introduced by Bersini et al.... In PAGE 11: ...291 DEC progressively removes 6 of the 20 initial rules. The results of Table1 and the observation of Fig. 8 reveals the high quality at the GTS-DEC-14 forecasts.... ..."

### Table 10. MSE obtained from applying each forecasting technique to the Nut Core time series

"... In PAGE 14: ... Using the results from the confidence interval, it is a sensible choice to try the first three techniques and compare which one does a better forecasting job. Table10 shows the results in terms of the real MSE, where the best three predicted techniques matched the best three techniques. ... ..."

### Table 5. Model Parameters: Country-Relative Forecast Model (time series/cross-sectional regression parameters; t-statistics in parentheses)

2002

Cited by 1

### Table 2. Combined forecast for the ACDC for the forthcoming 48 month. The time series values for the years 2004 to 2007 were not available at the time this study was conducted.

"... In PAGE 5: ... conducted their study. The forecasted atmospheric carbon dioxide levels are detailed for the next 4 years in Table2 . No data for the year 2004 was available at the time this study was conducted.... ..."

### Table 4. Forecasting 12 months ahead: Furnas time series Actual M. Likelihood Bayesian C.I. (95%)

"... In PAGE 6: ...ig. 4. Histograms for the parameters f 1, f 2 and t generated via MCMC simulation using prior density t- Gamma. Table4 shows the forecasted values for a period of 12 months ahead obtained for both the classical maximum likelihood approach and Bayesian method with a t-Gamma prior density. The forecasting uncertainty is quantified by the credible intervals shown in the fourth column of Table 4.... In PAGE 6: ... The forecasting uncertainty is quantified by the credible intervals shown in the fourth column of Table 4. The values in Table4 are in the original domain of streamflow measurements (and not in that of the transformed series Zt) obtained through the following inverse transformation: E(yf|f , t ) = exp{m m+ 2 m s [2E(Zf|f , t , Zp)+s mVar(Zf|f , t , Zp)]} b 1 f b 2 f b t b 1 f b 2 f b... In PAGE 7: ... Fig. 5 plots the forecasted values in Table4 for the Bayesian approach together with the actual streamflow values for a certain part of the Furnas time series. Fig.... ..."

### Table 1. Evaluation of 29 quarters out-of-sample forecasting performance for naive and unbiased forecasts from a VAR(3) model for log-transformed series for the untransformed level time series.

"... In PAGE 8: ... This means that the farther the time horizon the better is the performance of the unbiased forecast with respect to the naive. In Table 2 we report the same statistics as in Table1 but using only the forecasts from 10 to 29 periods ahead (20 forecasts). Hence, for the longer horizons one clearly needs the use of the unbiased forecasts.... ..."