Aggregation Bias Research Articles

Beyond the threshold: Understanding the asymmetric effects of renewable energy on CO2 emissions

While renewable energy deployment is essential to mitigate climate change, the interplay between renewable energy consumption and environmental degradation may not be linear. The environmental aspect of renewable energy consumption may change over time, depending on the scale and technique effects. This may be due to asymmetry in the relationship. Nonetheless, most current literature either assumes linearity, or ignores the turning point of the behavioral change. This results in inconclusive empirical findings at the disaggregated level of renewable energy consumption. This paper utilizes threshold estimation technique to capture the asymmetry in the renewable energy-CO2 emissions relation in the top ten renewable energy consumers covering the period 1990–2020. The literature gap is addressed by deriving the threshold effect at the aggregate and disaggregated levels to prevent aggregation bias. Understanding the thresholds of different renewable energy sources would improve policy effectiveness and resource allocation at different consumption levels to better curb climate change. The threshold estimation technique measures total renewables, hydro, solar, wind, and others (bioenergy and geothermal) as threshold variables. The findings indicate that total renewables and solar consumptions have stronger mitigating effects on CO2 emissions beyond the consumption levels of 4363.37 and 43.58 kWh, respectively. The advantageous environmental effect of wind consumption only manifests above the consumption level of 657.40 kWh. For policy implication, this study recommends an increase in the weightage of renewables in the energy mix by formulating energy-specific policies, in order to optimize the environmental benefits of renewable energy adoption.

Laboratory Data as a Potential Source of Bias in Healthcare Artificial Intelligence and Machine Learning Models.

Artificial intelligence (AI) and machine learning (ML) are anticipated to transform the practice of medicine. As one of the largest sources of digital data in healthcare, laboratory results can strongly influence AI and ML algorithms that require large sets of healthcare data for training. Embedded bias introduced into AI and ML models not only has disastrous consequences for quality of care but also may perpetuate and exacerbate health disparities. The lack of test harmonization, which is defined as the ability to produce comparable results and the same interpretation irrespective of the method or instrument platform used to produce the result, may introduce aggregation bias into algorithms with potential adverse outcomes for patients. Limited interoperability of laboratory results at the technical, syntactic, semantic, and organizational levels is a source of embedded bias that limits the accuracy and generalizability of algorithmic models. Population-specific issues, such as inadequate representation in clinical trials and inaccurate race attribution, not only affect the interpretation of laboratory results but also may perpetuate erroneous conclusions based on AI and ML models in the healthcare literature.

Vietnam's exports to Korea and the real exchange rate: Post-crisis evidence from the multiple threshold nonlinear autoregressive distributed lag model

In this study, the asymmetric effects of the real exchange rate on Vietnam's exports to Korea were investigated using the multiple threshold nonlinear autoregressive distributed lag (MTNARDL) model. As a result, the MTNARDL model exhibited enhanced estimation accuracy compared to the single-threshold nonlinear autoregressive distributed lag (NARDL) model. Furthermore, the aggregate bias problem was solved using individual export data for 16 major industries instead of relying on the overall export value data. The results revealed the substantial asymmetric impact of real exchange rate fluctuations in the short and long run. Fluctuations in the real exchange rate boosted Vietnam's exports to the Republic of Korea in the long run, with minimal negative influence. This indicates that the depreciation of the Vietnamese dong against the United States (US) dollar encouraged exports in some industries but limited exports in others. However, the short-term effect of the real exchange rate remains uncertain, as it could be either negative or positive depending on the industry and threshold level. The diverse impacts of the real exchange rate fluctuations on exports across industries were attributed to factors such as the capacity to absorb foreign exchange risks, product heterogeneity, and the ease of adjusting inputs and production for each sector. Therefore, these findings may help policymakers develop Vietnam-specific policies to address long- and short-term changes in the threshold levels of real exchange rate fluctuations.

Regional aggregation bias in the global computable general equilibrium model: Issues in assessing the economic impact of climate change and trade liberalization

This study examined the bias introduced by regional aggregation (RA) when assessing the effect of climate change and trade liberalization through computational experiments. The analysis used global computable general equilibrium models with different pre-aggregation levels, based on the GTAP10 database. The bias was quantified by comparing the effects between large-scale and aggregated models. The results revealed that RA bias comprised over 35% of estimated effects related to climate change and global trade liberalization; it reversed the change direction in the simulation and deviated changes beyond the variation range observed in the sensitivity analysis; and it increased in developing countries where the proportions of imports were high. Much RA bias is artificially created by aggregating countries with different initial shocks, which then spreads to other countries through a pseudo-trade structure. Therefore, adopting a disaggregated global model is crucial to improve the accuracy of climate change assessments with diverse shocks worldwide.

Alternative household water affordability metrics using water bill delinquency behavior

Rising water prices threaten affordable access to basic water service in the U.S., especially in low-income communities. Faced with unaffordable water bills, households may use less water than is healthy, forgo other essential services, or fall behind on water bill payments, risking water shutoffs. Affordability ratios (ARs), which express water bills as a fraction of income, are the most common measure of water affordability. However, ARs can underestimate unaffordability due to both spatial aggregation bias and their reliance on indirect proxies for ability to pay. New metrics are needed to identify households at risk of water insecurity due to affordability challenges. Here we investigate alternative water affordability metrics that use water bill late payments and debt to track actual payment behavior at the household level. We define metrics that capture the frequency, duration, and severity of water bill delinquency. We apply these metrics to a case study in Santa Cruz, California, using monthly billing data for approximately 40 000 households from 2009 through 2021. We find large variation in delinquency across households and over time, with higher delinquency linked to proxies for low wealth such as lower assessed home value. Census blocks with similar ARs often have distinct patterns of delinquency behavior, suggesting that block-level median affordability estimates may be masking sub-populations facing affordability challenges. These results highlight the benefits of using multiple, household-level metrics to capture the role affordability plays in household water security.

Aggregation Bias in the Measurement of US Global Value Chains

This paper employs US census microdata to construct establishment-level measures of global value chains (GVCs) and then assesses aggregation bias—the bias occurring when an entire industry is essentially treated as a single establishment. Our establishment-level measures indicate little slowdown in GVC expansion between 2002 and 2012. Related, the aggregation bias is negative, and it increased during this period. A decomposition reveals that most of the increase is from within establishments with high export and import intensities. Our results suggest that granular GVC measures will provide further understanding about how firms and economies adjust to shocks.

Experimental validation of an advanced impedance pneumography for monitoring ventilation volume during programmed cycling exercise

Objective. Impedance pneumography (IP) has provided static assessments of subjects’ breathing patterns in previous studies. Evaluating the feasibility and limitation of ambulatory IP based respiratory monitoring needs further investigation on clinically relevant exercise designs. The aim of this study was to evaluate the capacity of an advanced IP in ambulatory respiratory monitoring, and its predictive value in independent ventilatory capacity quantification during cardiopulmonary exercise testing (CPET). Approach. 35 volunteers were examined with the same calibration methodology and CPET exercise protocol comprising phases of rest, unloaded, incremental load, maximum load, recovery and further-recovery. In 3 or 4 deep breaths of calibration stage, thoracic impedance and criterion spirometric volume were simultaneously recorded to produce phase-specific prior calibration coefficients (CCs). The IP measurement during exercise protocol was converted by prior CCs to volume estimation curve and thus calculate minute ventilation (VE) independent from the spirometry approach. Main results. Across all measurements, the relative error of IP-derived VE (VER) and flowrate-derived VE (VEf) was less than 13.8%. In Bland-Altman plots, the aggregate VE estimation bias was statistically insignificant for all 3 phases with pedaling exercise and the discrepancy between VER and VEf fell within the 95% limits of agreement (95% LoA) for 34 or all subjects in each of all CPET phases. Significance. This work reinforces the independent use of IP as an accurate and robust alternative to flowmeter for applications in cycle ergometry CPET, which could significantly encourage the clinical use of IP and improve the convenience and comfort of CPET.

Estimating Market Power Exertion in the U.S. Beef Packing Industry: An Illustration of Data Aggregation Bias Using Simulated Data

This study investigates data aggregation bias in estimating market power in the U.S. beef packing industry using New Empirical Industrial Organization (NEIO) models and shows empirical procedures that can alleviate the bias. Unlike many earlier studies in estimating market power exertion, our study examines the data aggregation bias when market-level data are used in place of firm-level data and show how the bias could be reduced. We first derive data aggregation bias analytically, then empirically investigate the aggregation bias by estimating both firm and aggregate industry models. Because the firm-level data are not available, we use simulated data generated from the Monte Carlo simulation method. Hybrid models, combining limited firm-level data with aggregate data, are also estimated to illustrate how the aggregation bias could be reduced. Our results show that aggregate models with industry-level data tend to underestimate market power exertion in the U.S. beef packing industry, and the aggregation bias is statistically significant at the 1% level. Comparing results from hybrid models with firm-level estimates, we find that hybrid models reduce the bias but do not remove the aggregation bias significantly. The sensitivity analysis shows that market power estimate and aggregation bias are sensitive to functional forms.

Quantifying the spatial aggregation bias of urban heat data

Every year, high temperatures send people to the hospital and morgue, and the combination of climate change and urbanization will increase extreme heat exposure. Cities are searching for ways to determine the most affected areas to begin addressing this pervasive issue. While we are living through the “big data” revolution, policy makers are still uncertain about what level of data is most useful. We evaluate the data loss from using data at different spatial resolutions to evaluate heat vulnerability, as both the definition of intra-urban heat and the resolution of the data affect the area identified and targeted for mitigation. Variance-based metrics provide many advantages, but when data is aggregated, these metrics are less able to represent the full range of urban heat. Using the case of Bexar County (home to San Antonio, TX), we find that increasing data aggregation increases both false positive and false negative identification of intra-urban heat islands, leading to unreliable results. Misclassification increases as aggregation increases, indicating that decisions should be made at the finest spatial resolution possible.

Energy-related uncertainty and international stock market volatility

This paper predicts the daily return volatility of 28 advanced and developing stock markets using monthly metrics of the corresponding country and global energy-related uncertainty indexes (EUIs) recently proposed in the literature. Using data in their “natural” frequencies to avoid aggregation bias, the results show that country-specific and global EUIs have predictive powers for stock returns volatility for the in-sample periods, with increased levels of EUIs exhibiting the tendency to heighten volatility. This predictability also withstands various out-of-sample forecast horizons, implying that EUI is a statistically relevant predictor in the out-of-sample analysis. The forecast precision of the GARCH-MIDAS model is improved by incorporating global EUIs relatively more than country-specific EUIs. The robustness of the findings with respect to the choice of EUI and sample definition is further confirmed. The outcomes have important policy implications for the concerned stakeholders who are concerned with stability in the global financial system and economy.

An investigation into aggregation bias: the case of stocks and treasury bill returns in Ghana

An investigation into aggregation bias: the case of stocks and treasury bill returns in Ghana

Advancing unanchored simulated treatment comparisons: A novel implementation and simulation study.

Population-adjusted indirect comparisons, developed in the 2010s, enable comparisons between two treatments in different studies by balancing patient characteristics in the case where individual patient-level data (IPD) are available for only one study. Health technology assessment (HTA) bodies increasingly rely on these methods to inform funding decisions, typically using unanchored indirect comparisons (i.e., without a common comparator), due to the need to evaluate comparative efficacy and safety for single-arm trials. Unanchored matching-adjusted indirect comparison (MAIC) and unanchored simulated treatment comparison (STC) are currently the only two approaches available for population-adjusted indirect comparisons based on single-arm trials. However, there is a notable underutilisation of unanchored STC in HTA, largely due to a lack of understanding of its implementation. We therefore develop a novel way to implement unanchored STC by incorporating standardisation/marginalisation and the NORmal To Anything (NORTA) algorithm for sampling covariates. This methodology aims to derive a suitable marginal treatment effect without aggregation bias for HTA evaluations. We use a non-parametric bootstrap and propose separately calculating the standard error for the IPD study and the comparator study to ensure the appropriate quantification of the uncertainty associated with the estimated treatment effect. The performance of our proposed unanchored STC approach is evaluated through a comprehensive simulation study focused on binary outcomes. Our findings demonstrate that the proposed approach is asymptotically unbiased. We argue that unanchored STC should be considered when conducting unanchored indirect comparisons with single-arm studies, presenting a robust approach for HTA decision-making.

Trade policy and environmental quality in Africa: Aggregation issues and policy deductions

The significance of the impact of trade on sustainable development cannot be overstated. Even though there is a large amount of literature examining the trade - environment connection, most of the studies that have been done so have primarily examined variables like trade openness and foreign direct investment, ignoring the impact of particular trade policy tools, like tariffs. Moreover, there is a pertinent concern about aggregation bias in aggregate estimations, which could lead to inaccurate assessments of the overall connections and potentially misguide policymakers. In contrast to the prevailing trend, this study delves into the influence of trade policy, specifically tariffs, on the environmental degradation in 20 African nations. The investigation also addresses the issue of aggregation bias, aiming to provide a comprehensive understanding of the trade policy-environment link. The panel long-run estimates, employing the PMG estimator, reveal a significantly negative influence of tariffs on ecological footprints. This suggests that, in this context, tariffs do not create market inefficiency but rather contribute to strengthening environmental quality in the African sub-region. The findings underscore the importance of considering specific trade policy instruments in environmental policymaking. Policymakers in the region are presented with an opportunity to strategically utilize tariffs as a tool to address environmental degradation issues. The study further highlights that the relationship is negative at the panel level but positive at some of the individual country-level, indicating the presence of aggregation bias in estimating the trade policy-environmental dilapidation association on a panel scale. As aggregation bias may lead to misguided environmental policies, policymakers are prevailed on to carefully consider both panel and individual country-particular level when formulating trade and environmental strategies.

A Comparative Assessment and its Characterization of the Integrated Novel Water Pollution Index and its Statistical Approach for the Evaluation of Spatial Variations Using Factor Analysis: A Geospatial Approach in Mahanadi River, Odisha

Knowledge on water quality and its assessment, is necessary for both human health and environmental benefit. To account for spatial distribution, surface water quality parameters were analysed using integrated interpolation, geographical information systems (GIS) and multivariate analysis. A total of 19 locations and 13 water quality indicators were analysed, for a duration of six years (2018-2024). The study’s main objective was to assess the seasonal and regional variations in the water quality index (WQI) of Mahanadi River in Odisha using (N) pi, (S) pi, (O) pi, (C) pi, (E)y-WQI, Int w-WQI and Multivariate Statistical tools namely Factor Analysis (Fa). However, in the current investigation, pH, HCO3-, Na+, K+ and Mg2+ were within the permissible limits as per WHO standards. According to this study, the order of prevalence of ion concentrations is signified as follows: Mg2+ > Ca2+ > K+ > Na+ for cations and HCO3- > Cl- > SO42- for anions. The analysis of (N) pi indicated that about 15.79% of the sampled area, is affected by turbidity content, which is highly unsuitable for consumption. However, the remaining area (84.21%) is within the safe category of water. Classification of water based on (S) pi represents most of water samples falls between good water quality. Three unsuitable samples is noted as a result of excessive TDS and EC. In case of (O)pi, over 84.21% of the samples fell into categories of excellent, indicating the suitability for human activities. Using surface water quality results from (C) pi model, that reflects that out of 19 samples, 16 were suitable for drinking. Whereas 2 were polluted and 1 is seriously polluted, thus promotes unsuitability. Although there are several established techniques for calculating the WQI, the current study uses the quality index to consider a variety of water quality concerns in a cohesive manner. Meanwhile, in case of (E)y-WQI, 84.30% were excellent whereas 10% and 5% were poor and high polluted category. Over 42.11% of the samples fell into the categories of poor/very poor/not suitable, using the Int w-WQI diagram. Therefore, using these six approaches resembles a precise and comprehensive method to comprehend water quality in relation to pollution for human usage. In later stage, a factor analysis (Fa) can be applied to lessen the subjectivity and dimension of water quality characteristics. It reveals that the first five principal components explain almost 95.61% of dataset variation. This method removes the aggregation problems, weighting, opacity, and biases seen in traditional water quality evaluation techniques. The results of Fa suggested that turbidity, TKN, Ca2+ and Cl-, were the primary determinants of the water’s quality. The amount of organic pollution that was released into the river was influenced by anthropogenic activity in the vicinity of the river. In addition, the traditional dense habitation next to the river and the manufacturing waste that is transported from upstream to downstream are the sources of the high amount of TKN in urine and faeces. Therefore, given the high spatial distribution of geogenic turbidity and TKN occurrence, the study’s findings minimize uncertain causes and offer insights into surface water pollution regimes. They will also be useful to policy makers in helping to better plan, allocate resources, and manage the area’s potable water supply.

Teachers' judgment accuracy: A replication check by psychometric meta-analysis.

Teachers' judgment accuracy is a core competency in their daily business. Due to its importance, several meta-analyses have estimated how accurately teachers judge students' academic achievements by measuring teachers' judgment accuracy (i.e., the correlation between teachers' judgments of students' academic abilities and students' scores on achievement tests). In our study, we considered previous meta-analyses and updated these databases and the analytic combination of data using a psychometric meta-analysis to explain variations in results across studies. Our results demonstrate the importance of considering aggregation and publication bias as well as correcting for the most important artifacts (e.g., sampling and measurement error), but also that most studies fail to report the data needed for conducting a meta-analysis according to current best practices. We find that previous reviews have underestimated teachers' judgment accuracy and overestimated the variance in estimates of teachers' judgment accuracy across studies because at least 10% of this variance may be associated with common artifacts. We conclude that ignoring artifacts, as in classical meta-analysis, may lead one to erroneously conclude that moderator variables, instead of artifacts, explain any variation. We describe how online data repositories could improve the scientific process and the potential for using psychometric meta-analysis to synthesize results and assess replicability.

An extended combined compromise solution framework based on novel intuitionistic fuzzy distance measure and score function with applications in sustainable biomass crop selection

Now-a-days, due to the increasing demand of the energy resources across the world, choosing the best sustainable biomass crop for the production of biofuels is a strategic decision-making problem. This is generally accepted that intuitionistic fuzzy sets (IFSs) are much more efficient in comparison with fuzzy sets at representing and processing the uncertainty in real-life problems. Proceeding on the same line, this paper attempts to introduce an extended combined compromise solution (CoCoSo) framework to analyze the sustainable biomass crop selection (SBCS) problem in an intuitionistic fuzzy environment. In this framework, we suggest a new integration function based on double normalization multiple aggregation approach to overcome the aggregation biases of the original CoCoSo approach and discuss its advantages with some numerical examples. We also develop a combined weighting strategy based on distance measure and decision experts’ (DEs) opinions to evaluate the significance of criteria . For this, we propose a novel distance measure (DM) and establish its superiority through some numerical comparisons. Also, the rationality of the suggested measure over the extant measures is justified by the use of an algorithm based on the developed measure for pattern recognition issues. In this framework, the comparison issue of IFSs is resolved by proposing a new score function. Furthermore, a case study of the SBCS is presented for the implementation of the developed CoCoSo approach, which confirms the viability and effectiveness of the new methodology. The results of the sensitivity analysis demonstrate that option “Miscanthus” consistently achieves the highest rank and is independent of variations of trade-off parameter and balancing factor. Finally, a comprehensive comparison is carried out to ensure the steadiness and reliability of the introduced framework.

Respondent biases in agricultural household surveys

Two sources of respondent bias introduce measurement error into household statistics: asymmetric information between the proxy respondent and the individual on whom they report; and aggregation bias when a proxy respondent reports on a household-level outcome across multiple individuals. We estimate the effects of respondent biases in a survey experiment in Burkina Faso by varying who reports on the agricultural production of household members. We find respondent biases are not solely attributable to asymmetric information. Choosing a household head proxy lowers aggregation biases, but results in both over and under-estimates of agricultural variables relative to random proxies. Random proxies systematically under report agricultural statistics. Self-reporting protocols increase enumerator work days by only 5% indicating a high bias-cost tradeoff in choosing proxy response over self-reports. Survey designers should weight whether proxy bias magnitude or direction of bias are more significant threats to parameter estimation when determining their proxy response protocol.

Intertemporal budget constraint and current account sustainability: Evidence from a large emerging economy

This paper examines the sustainability of the current account deficit (CAD) and validity of the intertemporal budget constraint of India for the period of 1980 to 2019. The cointegration results show that exports and imports are not cointegrated and the current account (CA) series is not mean reverting during the study period. Therefore, the results refute the sustainability of CAD and the validity of the intertemporal budget constraint in the Indian context. Furthermore, the study also emphasised the testing of the disaggregated CA and found that while the goods account is not sustainable, the services account found to be sustainable in the long run. Our findings thus indicate the potential for aggregation bias to have an impact on the overall CA sustainability results. Therefore, policies that help in achieving self‐sufficiency on the one hand and export promotion in goods on the other may eventually contribute to CA sustainability.

Are gross financial inflows expansionary or contractionary? Evidence from emerging economies

This paper put to the test two hypotheses. First, do gross financial inflows lead to an expansionary or contractionary effect in emerging economies? Second, do different financial inflows exhibit diverse impacts on domestic output and if there is any aggregation bias? However, our work ameliorates the existing literature by tackling the problem of slope heterogeneity and cross-correlated errors by adopting dynamic common-correlated predictors. The analysis further identify different regimes in the regression relationships, considering our sample includes significant global financial imbalances, global financial crisis, taper tantrum period, etc. We find that all types of gross financial inflows, except gross debt inflows, are expansionary. However, there is evidence of possible aggregation bias in using aggregated gross financial flows. Our findings imply that a distinction between different types of financial inflows is necessary to make informed policy recommendations since policies based on aggregate inflows may be ill-advised. Further, emerging markets should focus on encouraging direct and portfolio equity inflows and improving local financial markets’ absorptive capacities in borrower countries to improve welfare gains.

Modelling the continuum of serious traffic injuries in police-hospital linked data by applying the random parameters hazard-based duration model

Injury severity in police crash reports is usually recorded in three to five classes, including property damage, slight, moderate, serious, and fatal injuries. Among these classifications, serious injuries are commonly classified as cases where a road user is admitted to a hospital. In this classification, the length of hospital stay is not differentiated, whether one day or ten days, as long as the road user has been admitted to the hospital. As such, the inferences drawn from assuming that all serious injuries (1 if a road user is admitted to the hospital; 0 otherwise) are at the same severity level inherently suffer from aggregation bias and may not provide a thorough understanding of this severity category. This study proposes a hazard-based duration modelling approach to examine the severity of serious injury crashes measured in a continuous spectrum. Specifically, using the length of hospital stay as the measure of serious injuries, a random parameters hazard-based duration model with heterogeneity in means was applied to model serious injury crashes obtained by linking crash records in police and hospital databases. To address temporal instability, the injury records sources from Abu Dhabi, United Arab Emirates (UAE), between 2015 and 2019 were modelled separately for each year. The results showed that factors positively associated with more serious injury severity (prolonged length of hospital stay) are rural areas, high posted speed limits of 100–160 km/h, overturned crashes, speeding, impaired driving, involvements of a heavy vehicle, nighttime crashes, lack of restraint usage, and injuries to the head or lower extremities. In particular, speeding violations during nighttime are positively associated with more serious injuries. Furthermore, the means of the random parameters of head injury are positively influenced by speeding, lack of restraint usage, and motorcycle involvement through the heterogeneity-in-mean specification of the hazard-based duration model. The proposed modelling approach to model serious traffic injuries using a hazard-based duration model provides a comprehensive understanding of the factors associated with serious injuries.