### Table 29. Effect of Weather Forecast on Building Load Forecast

in NOTICE

"... In PAGE 11: ...able 28. Accuracy of Weather Forecast.................................................................................99 Table29 .... ..."

### Table 2. Summary Statistics for Consensus Forecast Errors

"... In PAGE 10: ... These plots show clearly the underprediction of inflation and the interest rate, and the relative accuracy of the unemployment rate consensus forecasts. Excepting the GDP growth forecasts, the pattern of standard deviations rising with the forecast horizon is again a feature of the errors (see Table2 ); this increase in volatility can be seen in the plots, particularly for the T-bill rate and unemployment rate. The figures also highlight greater volatility exhibited in the early 1980s for most of the series.... In PAGE 12: ... For a given variable and horizon, cross-section summary statistics are calculated using all the available forecasters at a given point in time; these are then averaged over time, yielding the mean summary statistics (and the standard errors associated with these means) presented in Table 4. The average of the cross-section means for each variable and horizon is the same as the mean of the consensus forecast error in Table2 ; these results have already been discussed but are included here for completeness. The number of actuals in each case, i.... ..."

### Table 5: Point Forecast Evaluation: Multiple Forecast Encom- passing Test

"... In PAGE 15: ... The MSMH model seems to be second-best at the shorter forecast horizons whereas the SETAR model takes the second place at the longer forecast horizons. As seen from Table5 , the null hypothe- sis that the AR model forecast encompasses all other models simultaneously is rarely rejected. However, the same is true for the MSMH and MSIAH models and for the SETAR model at longer forecast horizons.... ..."

### Table 3 Probabilistic flood forecast data t = 1 t = 2 t = 3

"... In PAGE 12: ...associated with disrupting and evacuating an individual is assumed to be 10,000 yen, the average value associated with each human hour lost due to the evacuation is assumed to be 1000 yen, and the value associated with a human life is set at 50,000,000 yen. Probabilistic flood forecast data for 1, 2 and 3-hour ahead forecasts made for Mino at hourly steps between midday and 15:00 are given in Table3 for a hypothetical event. Although the example given considers only three forecast periods, the use of a 6-hour ahead forecast would be used in the same manner.... ..."

### Table 2. Summary of probabilistic beliefs and correct forecasts Variable

"... In PAGE 8: ... Statistical summary for these probabilistic beliefs vis. correctness of the respective forecasts on a firm-by-firm basis are summarized in Table2 for all periods (deviations across periods, as well as across industries and regions, were most often statistically in- significant for all reported indicators). The upper part of the table reports the main summary statistics; it clearly implies that the managers were quite confi- dent in their forecasts: the average probabilistic belief varied across indicators from 74.... ..."

### Table 2 gives the peak load forecasting results in

### Table 2: Results for Utility Load Forecasting

"... In PAGE 21: ...vailable before 10/1/91, i.e., the weekdays data from 1/1/85 to 9/30/91. To test the performance of the predictions, in Table2 we report three measures that were used in [6, 7]. The rst measure is the Mean Absolute Percentage Error (MAPE).... In PAGE 22: ...Table2 contains MAPE, CV and Median CV of the neural network prediction gb (X; Z). This re ects the prediction error level achieved by this neural network model.... In PAGE 23: ... To serve as a benchmark, the prediction error levels for the winner QUERI model are reported in the fourth row of Table 2. From the rst row to the second row of Table2 , we see that the replace- ment of X by its noisy estimators W increases the prediction error. However, not all the increases in the prediction error are intrinsically due to the usage of noisy predictors.... In PAGE 24: ... As mentioned earlier, the proposed procedure is only adjusting the prediction for the use of noisy surrogate predictors. Hence by itself it won apos;t be able to improve the prediction beyond the neural network prediction obtainable knowing the true predictors as reported the rst row of Table2 . If we want to beat the QUERI model, then we need to put in extra modeling e ort for the neural network model [7].... ..."

### Table 4 Point Forecast Accuracy Comparisons

2005

"... In PAGE 13: ... This ensures a fair apples-to-apples comparison. We report RMSPEs in Table4... In PAGE 28: ...Table4 (Continued) Point Forecast Accuracy Comparisons Daily Average Temperature Persistence Climatological Autoregressive EarthSat Las Vegas 1-Day-Ahead 3.78 5.... ..."

Cited by 4