Forecasting Peak Daily Ozone Levels: Part 2—A Regression with Time Series Errors Model Having a Principal Component Trigger to Forecast 1999 and 2002 Ozone Levels

Liu, Pao-Wen Grace; Johnson, Richard A.

doi:10.1080/10473289.2003.10466321

Cited by 10 publications

(9 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The superiority of multivariate over univariate time series models has been underscored by many researchers. 5,31,40,47,49,51 However, for simulating the days of ozone greater than 120 ppb, the RTSE model cannot compete with the RTSE model that includes a PC variable. This is especially true when the simulation profiles of 1998 and 2001 in Figures 5 and 6 are examined.…”

Section: Resultsmentioning

confidence: 99%

“…It was found useful in applying a welldesigned ozone simulation model to forecast the peak daily 1-hr ozone concentration. 5,40 However, the performance of the real-time forecasting was not as accurate as that of the simulation because of inclusion of errors from weather forecasts, such as forecasted temperature, wind speed, and wind direction. On the other hand, the conclusions that a multivariate time series model performed better than that of a univariate time series model, and the multivariate model containing a PC variable is superior to the same model without the PC variable, were true regardless of the RTSE models used for ozone simulation or forecast.…”

Section: Pod ϭmentioning

confidence: 99%

“…On the other hand, the conclusions that a multivariate time series model performed better than that of a univariate time series model, and the multivariate model containing a PC variable is superior to the same model without the PC variable, were true regardless of the RTSE models used for ozone simulation or forecast. 40 …”

Section: Pod ϭmentioning

confidence: 99%

“…assumption). 39,40 Pearson correlations and t tests provide meaningful results only if the i.i.d assumption can be satisfied. However, the assumption of statistical independence is violated because of serial or autocorrelation.…”

Section: Investigation Of Ozone-related Variablesmentioning

confidence: 99%

“…8 In other words, the ozone exceedances have been overwhelming at Ta-Liao, and the peak daily 1-hr ozone values have gradually replaced PM 10 as the major pollution indicator for the PSI above 100 warning. 8 Unlike the mid-latitude cities in the United States (35)(36)(37)(38)(39)(40)(41)(42)(43)(44)(45) o N), which encounter only one single ozone season (summer) every year, in Taiwan, ozone exceedances have been observed in two seasons-spring and late autumn. 7 As a result, the ozone problem in Taiwan has been concerned not only with the remarkable high values but also with the prolonged ozone seasons that frequent exceedances were observed.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Establishment of a Box-Jenkins Multivariate Time-Series Model to Simulate Ground-Level Peak Daily One-Hour Ozone Concentrations at Ta-Liao in Taiwan

Liu

2007

Journal of the Air & Waste Management Association

Self Cite

View full text Add to dashboard Cite

Box-Jenkins univariate autoregressive integrated moving average (ARIMA) and regression with time-series error (RTSE) models were established to simulate historical peak daily 1-hr ozone concentrations at Ta-Liao, Taiwan, 1997-2001. During 1995, the 600 days of Pollution Standard Index (PSI) more than 100 (peak daily 1-hr ozone concentrations detected by greater than 120 ppm) at TaoLiao showed the highest ozone exceedances among the six monitoring stations in Kaohsiung County. To improve the predictability of extremely high ozone, two different principal components, PC1 and PC(1 ϩ 2), were introduced in the RTSE model. Four typical predictors (particular matter with an aerodynamic diameter less than or equal to 10 m, temperature, wind speed, and wind direction) plus a PC trigger remained significant in the RTSE model. The model performance statistics concluded that the RTSE model with PC1 was optimal, compared with the univariate ARIMA, the RTSE model without PC, and RTSE model with PC(1 ϩ 2). The contingency table shows that the successful predictions of the univariate model were only 12.9% of that of the RTSE model with PC1. Also, the POD value was improved approximately 5-fold when the univariate model was replaced by the RTSE model, and almost 8-fold when it was replaced by the RTSE model with PC1. Moreover, introducing the PC trigger indeed enhanced the ozone predictability. After the PC trigger was introduced in the RTSE model, the POD was increased 69.9%, and the FAR was reduced 8.3%. The overall correlation between the observed and simulated ozone was improved 9.6%. Also, the first principal component was more useful than the first two components in playing the "trigger" role, though it counted only for 58.62% of the environmental variance during the high ozone days.

show abstract