Use Of Real-time Data Research Articles

Use of real-time data in environmental monitoring: current practices

Water quality monitoring in Europe, especially in transboundary water courses has made steady progress during the last decades through establishment of international commissions. The main activities of these commissions include protection and management of the catchment, sustainable use of the river and establishment of Accident Emergency Warning Systems (AEWS). The latter could be effectively accomplished only through real-time monitoring. Concurrently real-time data have been found important for the monitoring of potable water intake points, wastewater treatment plants, estuaries and in aquaculture. With the recognition of the diversified demand, there are a number of questions to be answered such as: (1) are we satisfied with the existing monitoring systems? (2) is standardisation of the measuring instruments a necessity? (3) do we have foolproof systems for data capture and transmission? (4) are there adequate procedures to analyse vast amount of data generated? We have to answer these questions urgently as the demand for real-time monitoring has been drastically increased.

Forecasting with a real-time data set for macroeconomists

This paper discusses how forecasts are affected by the use of real-time data rather than latest-available data. The key issue is this: in the literature on developing forecasting models, new models are put together based on the results they yield using the data set available to the model’s developer. But those are not the data that were available to a forecaster in real time. How much difference does the vintage of the data make for such forecasts? We explore this issue with a variety of exercises designed to answer this question. In particular, we find that the use of real-time data matters for some forecasting issues but not for others. It matters for choosing lag length in a univariate context. Preliminary evidence suggests that the span––or number––of forecast observations used to evaluate models may also be critical: we find that standard measures of forecast accuracy can be vintage-sensitive when constructed on the short spans (five years of quarterly data) of data sometimes used by researchers for forecast evaluation. The differences between using real time and latest-available data may depend on what is being used as the “actual” or realization, and we explore several alternatives that can be used. Perhaps of most importance, we show that measures of forecast error, such as root-mean-squared error and mean absolute error, can be deceptively lower when using latest-available data rather than real-time data. Thus, for purposes such as modeling expectations or evaluating forecast errors of survey data, the use of latest-available data is questionable; comparisons between the forecasts generated from new models and benchmark forecasts, generated in real time, should be based on real-time data.

Reply to the comments on ‘The state of macroeconomic forecasting’

Given the compliments that our commentators have made on our survey, we cannot possibly have many arguments with them. We want to thank them for all their efforts and the way in which they have elaborated on our work and provided new ideas that we had not considered. These comments were especially welcome because all of them have made major contributions to the field of forecasting either in developing the statistical and theoretical foundations of macroeconomic forecasts or in administering and evaluating important forecasting surveys. We reply only to indicate where we basically agree and to note any major disagreements with their comments. The major thrust of these positive comments is that they indicate the directions that research should take in order to improve forecasting performance. Michael Clements shows that the theory of forecasting is well founded and provides theoretical justifications for the judgmental adjustments that forecast evaluations have found to be effective in reducing forecast errors. Forecast failures occur when unforeseen events occur. After the event, the model is set back on course through judgmental adjustment. The question, however, is: How do you predict these events in advance? A limited answer to this comes from the more extensive use of real-time data (Dean Croushore). Clements suggests that indicators might be combined with models. Arnold Zellners comments expand on this approach of melding indicators and models. He also cites many additional articles in which he and his co-authors have developed some of the ideas that we surveyed. Other common themes are also present in these comments. Zellner and Croushore focus on data problems to which we alluded, and Croushore and Kenneth West also suggest that more statistical testing be undertaken, a view we also share. Finally, Croushore and Ron Smith discuss the role that forecast users play and that the

Comments on ‘The state of macroeconomic forecasting’

The paper by Fildes and Stekler provides an overview of empirical results in macroeconomic forecasting. They do not address three key issues: (1) the importance of data revisions and the use of real-time data; (2) the statistical significance of differences between forecasts; and (3) the role of macroeconomic theory in formulating and evaluating macroeconomic forecasts.

Let's get “real” about using economic data

We show that using data which are properly available in real time when assessing the sensitivity of asset prices to economic news leads to different empirical findings than when data availability and timing issues are ignored. We do this using the framework of Chen et al. [J. Bus., 59 (1986)], and examine whether innovations to economic variables can be viewed as risks that are rewarded in asset markets. Our findings support the view that data uncertainty is sufficiently prevalent to warrant careful use of real-time data when forming real-time news measures, and in general when undertaking empirical financial investigations involving macroeconomic data.

Forecasting with a Real-Time Data Set for Macroeconomists

This paper discusses how forecasts are affected by the use of real-time data rather than latest-available data. The key issue is this: In the literature on developing forecasting models, new models are put together based on the results they yield using the data set available to the model's developer. But those are not the data that were available to a forecaster in real time. How much difference does the vintage of the data make for such forecasts? The authors explore this issue with a variety of exercises designed to answer this question. In particular, the authors find that the use of real-time data matters for some forecasting issues but not for others. It matters for choosing lag length in a univariate context. Preliminary evidence suggests that the span--or number--of forecast observations used to evaluate models may also be critical: the authors find that standard measures of forecast accuracy can be vintage-sensitive when constructed on the short spans (five years of quarterly data) of data sometimes used by researchers for forecast evaluation. The differences between using real-time and latest-available data may depend on what is being used as the actual or realization, and the authors explore several alternatives that can be used. Perhaps of most importance, the authors show that measures of forecast error, such as root-mean-squared error and mean absolute error, can be deceptively lower when using latest-available data rather than real-time data. Thus, for purposes such as modeling expectations or evaluating forecast errors of survey data, the use of latest-available data is questionable; comparisons between the forecasts generated from new models and benchmark forecasts, generated in real time, should be based on real-time data.

Let's Get 'Real' About Using Economic Data

We show that using data which are properly available in real time when assessing the sensitivity of asset prices to economic news leads to different empirical findings than when data availability and timing issues are ignored. We do this by focusing on a particular example, namely Chen, Roll and Ross (1986), and examine whether innovations to economic variables can be viewed as risks that are rewarded in asset markets. Our findings support the view that data uncertainty is sufficiently prevalent to warrant careful use of real-time data when forming real-time news measures, and in general when undertaking empirical financial investigations involving macroeconomic data.

Incorporating Crash Risk in Selecting Congestion-Mitigation Strategies: Hampton Roads Area (Virginia) Case Study

The extensive research under way to forecast traffic volumes from realtime data is expected to result in a significant increase in the use of realtime data for congestion mitigation and management in the next few years. Unfortunately, congestion-mitigation strategies based solely on these forecast volumes will disregard the impact of other traffic variables on crash occurrence. This may result in congestion-mitigation strategies that improve traffic flow but significantly increase crash rates. These other variables, for which real-time data are also available in traffic management centers, include speed and occupancy. However, real-time data have not been widely used in research to develop mathematical models that relate crash rates with these other traffic variables. The availability of this type of relationship will enable engineers in the field to develop congestion-mitigation strategies based not only on forecast volumes but also on real-time information on traffic variables such as speed and occupancy. A procedure that uses real-time data on traffic flow, speed, and occupancy and the relationship between these variables and crash occurrence could be used to develop congestion-mitigation strategies that incorporate safety. Currently, no such decision-support tool is readily available to the traffic engineer. The feasibility of developing such a methodology in which real-time data could be used to develop congestion-mitigation strategies that also consider crash risk was investigated. Models showing the relationship between flow and occupancy and between number of crashes and occupancy were developed for specific sites in the Hampton Roads, Virginia, area. These models were then used to illustrate how diversion strategies that consider crash risk may be developed.

Educational affairs

The EARTHSTORM Project is a K–12 teacher enhancement program for Oklahoma science and mathematics teachers who desire to incorporate real-time environmental data into their classroom activities. EARTHSTORM provides a model for the professional development of teachers, daily staff support for participants, scientific mentorship of students, development of data analysis tools for students and teachers, and use of real-time data in the K–12 classroom.

Managing load data at Virginia Power

The use of real-time data and what is referred to as semi-real-time data for asset management is described. The basic premise behind asset management is to utilize existing facilities to full capacity before installing new ones. Facility load data is made readily available to various company departments (such as metering, substation band system protection, budgeting, marketing, system planning, and division planning) as well as system and division operations by tying information networks together. Having access to this data allows unique approaches to reliability and efficiency in serving customers.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Contamination of soil and groundwater by automatic transmission fluid: Site description and problem assessment

Soil and groundwater beneath a region of a manufacturing plant are contaminated with automatic transmission fluid (ATF). The extent of contamination was assessed by maximizing the use of real-time data from soil-core sampling and monitoring wells. The number, location, and depth of cores and of monitoring wells were determined during the investigation based on: (1) inspection and analysis of soil-core samples immediately after each core was taken; (2) physical and chemical measurements of core samples at the end of each day; (3) measurements in monitoring wells at several stages during the investigation. This approach differs significantly from the conventional approach of randomly placing wells through the hydrogeologic system. Soil cores were taken and monitoring wells installed at 53 locations. The perched aquifer extends to about 13 ft. and is comprised mainly of sandy materials, which have spatial heterogeneity in size distribution and hydraulic properties. About 208 000 ± 33 000 gal. of ATF has spread over an area of about 64 000 ft. 2. The region of ATF contamination is comprised of three distinct and contiguous layers. The center layer is about 2.6ft. deep at its thickest point and extends to about 250ft. at its widest point. The soil in this zone is about 85% saturated with 133 000 ± 21 000 gal. of ATF, which has depressed the water table into the aquifer. The top layer is about 14in. thick and econtains about 50640 gal. of ATF held by capillary forces. The amount of ATF in this zone decreases with height above the center layer from about 85% saturation to residual saturation (20%). The amount of ATF in the deepest layer is near the residual saturation. This layer is 1.0 ± 0.5 ft. thick and has 24 500 ± 12 250 gal. ATF. This investigative approach did not spread the ATF to clean regions of the aquifer as could occur with conventional approaches, and it provided the data needed to assess the problem and to design a cleanup plan. A new approach is being used to recover the free ATF at the site.