在黄金期货价格预测问题的研究中,价格具有时变性、非线性、高噪声和影响因子复杂等因素,决定了其被准确预测的难度。传统方法对黄金期货价格的预测主要借助于静态模型,导致预测精度不高或分析不足。为了能动态而准确的预测黄金期货价格...在黄金期货价格预测问题的研究中,价格具有时变性、非线性、高噪声和影响因子复杂等因素,决定了其被准确预测的难度。传统方法对黄金期货价格的预测主要借助于静态模型,导致预测精度不高或分析不足。为了能动态而准确的预测黄金期货价格,本文从技术行情指标、行业方面的影响因素及宏观经济环境指标三个维度选取39个变量,以机器学习(machine learning;ML)方法构建基本融合素材,利用动态模型平均(dynamic model averaging,DMA)方法代替传统模型融合技巧,得到黄金期货价格预测模型。实证结果表明,采用机器学习-动态模型平均策略能够明显提高黄金期货价格的预测精度。展开更多
Three wells in New Hampshire were sampled bimonthly over three years to evaluate the temporal variability of arsenic concentrations and groundwater age.All samples had measurable concentrations of arsenic throughout t...Three wells in New Hampshire were sampled bimonthly over three years to evaluate the temporal variability of arsenic concentrations and groundwater age.All samples had measurable concentrations of arsenic throughout the entire sampling period and concentrations in individual wells had a mean variation of more than 7 μg/L.The time series data from this sampling effort showed that arsenic concentrations ranged from a median of 4 mg/L in a glacial aquifer well (SGW-65) to medians of 19 μg/L and 37 μg/L in wells (SGW-93 and KFW-87) screened in the bedrock aquifer,respectively.These high arsenic concentrations were associated with the consistently high pH (median ≥- 8) and low dissolved oxygen (median <0.1 mg/L) in the bedrock aquifer wells,which is typical of fractured crystalline bedrock aquifers in New Hampshire.Groundwater from the glacial aquifer often has high dissolved oxygen,but in this case was consistently low.The pH also is generally acidic in the glacial aquifer but in this case was slightly alkaline (median =7.5).Also,sorption sites may be more abundant in glacial aquifer deposits than in fractured bedrock which may contribute to lower arsenic concentrations.Mean groundwater ages were less than 50 years old in all three wells and correlated with conservative tracer concentrations,such as chloride;however,mean age was not directly correlated with arsenic concentrations.Arsenic concentrations at KFW-87 did correlate with water levels,in addition,there was a seasonal pattern,which suggests that either the timing of or multiple sampling efforts may be important to define the full range of arsenic concentrations in domestic bedrock wells.Since geochemically reduced conditions and alkaline pHs are common to both bedrock and glacial aquifer wells in this study,groundwater age correlates less strongly with arsenic concentrations than geochemical conditions.There also is evidence of direct hydraulic connection between the glacial and bedrock aquifers,which can influence arsenic concentrations.Correlations bet展开更多
Traffic congestion has caused many detrimental effects including higher fuel consumption, more vehicle emissions, increased accidents, as well as greater tension due to uncertain travel time. In addition to delay, the...Traffic congestion has caused many detrimental effects including higher fuel consumption, more vehicle emissions, increased accidents, as well as greater tension due to uncertain travel time. In addition to delay, the variability and reliability of travel time has been of concern to motorists on their daily travel, especially during peak periods. The objective of this study is to examine freeway travel time variability and reliability under different traffic and weather conditions with the use of TRANSMIT data collected by roadside readers deployed on a 40-mile segment of the Interstate Highway 1-287 in New Jersey. Travel time variability and reliability measures including mean travel time, the 95th percentile travel time, travel time index, buffer index, and planning time index under recurring and non-recurring congestion （adverse weather） are investigated. It was found that the standard deviation of travel time increased, due to the weather condition varying from dry to rain and to snow, while the buffer index increased from 29% to 45% and to 94%.展开更多
文摘在黄金期货价格预测问题的研究中,价格具有时变性、非线性、高噪声和影响因子复杂等因素,决定了其被准确预测的难度。传统方法对黄金期货价格的预测主要借助于静态模型,导致预测精度不高或分析不足。为了能动态而准确的预测黄金期货价格,本文从技术行情指标、行业方面的影响因素及宏观经济环境指标三个维度选取39个变量,以机器学习(machine learning;ML)方法构建基本融合素材,利用动态模型平均(dynamic model averaging,DMA)方法代替传统模型融合技巧,得到黄金期货价格预测模型。实证结果表明,采用机器学习-动态模型平均策略能够明显提高黄金期货价格的预测精度。
基金supported by the U.S.Geological Survey’s National Water-Quality Assessment Project
文摘Three wells in New Hampshire were sampled bimonthly over three years to evaluate the temporal variability of arsenic concentrations and groundwater age.All samples had measurable concentrations of arsenic throughout the entire sampling period and concentrations in individual wells had a mean variation of more than 7 μg/L.The time series data from this sampling effort showed that arsenic concentrations ranged from a median of 4 mg/L in a glacial aquifer well (SGW-65) to medians of 19 μg/L and 37 μg/L in wells (SGW-93 and KFW-87) screened in the bedrock aquifer,respectively.These high arsenic concentrations were associated with the consistently high pH (median ≥- 8) and low dissolved oxygen (median <0.1 mg/L) in the bedrock aquifer wells,which is typical of fractured crystalline bedrock aquifers in New Hampshire.Groundwater from the glacial aquifer often has high dissolved oxygen,but in this case was consistently low.The pH also is generally acidic in the glacial aquifer but in this case was slightly alkaline (median =7.5).Also,sorption sites may be more abundant in glacial aquifer deposits than in fractured bedrock which may contribute to lower arsenic concentrations.Mean groundwater ages were less than 50 years old in all three wells and correlated with conservative tracer concentrations,such as chloride;however,mean age was not directly correlated with arsenic concentrations.Arsenic concentrations at KFW-87 did correlate with water levels,in addition,there was a seasonal pattern,which suggests that either the timing of or multiple sampling efforts may be important to define the full range of arsenic concentrations in domestic bedrock wells.Since geochemically reduced conditions and alkaline pHs are common to both bedrock and glacial aquifer wells in this study,groundwater age correlates less strongly with arsenic concentrations than geochemical conditions.There also is evidence of direct hydraulic connection between the glacial and bedrock aquifers,which can influence arsenic concentrations.Correlations bet
文摘Traffic congestion has caused many detrimental effects including higher fuel consumption, more vehicle emissions, increased accidents, as well as greater tension due to uncertain travel time. In addition to delay, the variability and reliability of travel time has been of concern to motorists on their daily travel, especially during peak periods. The objective of this study is to examine freeway travel time variability and reliability under different traffic and weather conditions with the use of TRANSMIT data collected by roadside readers deployed on a 40-mile segment of the Interstate Highway 1-287 in New Jersey. Travel time variability and reliability measures including mean travel time, the 95th percentile travel time, travel time index, buffer index, and planning time index under recurring and non-recurring congestion （adverse weather） are investigated. It was found that the standard deviation of travel time increased, due to the weather condition varying from dry to rain and to snow, while the buffer index increased from 29% to 45% and to 94%.
