According to the Energy Information Administration, average retail gasoline prices tend to typically be higher in certain states than in others. Aside from taxes, the factors shown to contribute to regional and even l...According to the Energy Information Administration, average retail gasoline prices tend to typically be higher in certain states than in others. Aside from taxes, the factors shown to contribute to regional and even local differences in gasoline prices include proximity of supply, supply disruptions, competition in the local market and environmental programs. Of interest in this paper is proximity of supply. It has been hypothesized that areas farthest from the Gulf Coast （the source of nearly half of the gasoline produced in the United States and, thus, a major supplier to the rest of the country） tend to have higher prices. To test this hypothesis, the paper assembles state level monthly retail gasoline data for the period 1983 to 2007 for five states with oil refineries （Alabama, Georgia, Texas, Mississippi and Louisiana） and five states without refineries （Arkansas, Tennessee, North Carolina, South Carolina and Florida）. The analysis employs dynamic correlation, regression, cointegration and vector autoregressive methods. Overall, the results show that retail gas prices in states with refineries and those without refineries tend to move in the same direction over time. The small differences observed over time may suggest that price shocks take a short time to be felt nationwide.展开更多
We integrate k-Nearest Neighbors(kNN) into Support Vector Machine(SVM) and create a new method called SVM-kNN.SVM-kNN strengthens the generalization ability of SVM and apply kNN to correct some forecast errors of SVM ...We integrate k-Nearest Neighbors(kNN) into Support Vector Machine(SVM) and create a new method called SVM-kNN.SVM-kNN strengthens the generalization ability of SVM and apply kNN to correct some forecast errors of SVM and improve the forecast accuracy.In addition,it can give the prediction probability of any quasar candidate through counting the nearest neighbors of that candidate which is produced by kNN.Applying photometric data of stars and quasars with spectral classification from SDSS DR7 and considering limiting magnitude error is less than 0.1,SVM-kNN and SVM reach much higher performance that all the classification metrics of quasar selection are above 97.0%.Apparently,the performance of SVM-kNN has slighter improvement than that of SVM.Therefore SVM-kNN is such a competitive and promising approach that can be used to construct the targeting catalogue of quasar candidates for large sky surveys.展开更多
文摘According to the Energy Information Administration, average retail gasoline prices tend to typically be higher in certain states than in others. Aside from taxes, the factors shown to contribute to regional and even local differences in gasoline prices include proximity of supply, supply disruptions, competition in the local market and environmental programs. Of interest in this paper is proximity of supply. It has been hypothesized that areas farthest from the Gulf Coast （the source of nearly half of the gasoline produced in the United States and, thus, a major supplier to the rest of the country） tend to have higher prices. To test this hypothesis, the paper assembles state level monthly retail gasoline data for the period 1983 to 2007 for five states with oil refineries （Alabama, Georgia, Texas, Mississippi and Louisiana） and five states without refineries （Arkansas, Tennessee, North Carolina, South Carolina and Florida）. The analysis employs dynamic correlation, regression, cointegration and vector autoregressive methods. Overall, the results show that retail gas prices in states with refineries and those without refineries tend to move in the same direction over time. The small differences observed over time may suggest that price shocks take a short time to be felt nationwide.
基金supported by the National Natural Science Foundation of China(Grant Nos.10778724,11178021 and 11033001)the Natural Science Foundation of Education Department of Hebei Province (Grant No.ZD2010127)the Young Researcher Grant of National Astronomical Observatories,Chinese Academy of Sciences
文摘We integrate k-Nearest Neighbors(kNN) into Support Vector Machine(SVM) and create a new method called SVM-kNN.SVM-kNN strengthens the generalization ability of SVM and apply kNN to correct some forecast errors of SVM and improve the forecast accuracy.In addition,it can give the prediction probability of any quasar candidate through counting the nearest neighbors of that candidate which is produced by kNN.Applying photometric data of stars and quasars with spectral classification from SDSS DR7 and considering limiting magnitude error is less than 0.1,SVM-kNN and SVM reach much higher performance that all the classification metrics of quasar selection are above 97.0%.Apparently,the performance of SVM-kNN has slighter improvement than that of SVM.Therefore SVM-kNN is such a competitive and promising approach that can be used to construct the targeting catalogue of quasar candidates for large sky surveys.
