Model structure or data aggregation level: Which leads to greater bias of results?

Brockmeier, Martina; Bektasoglu, Beyhan

doi:10.1016/j.econmod.2014.01.003

Cited by 15 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Aggregation bias is also present in structural models, as shown in Foroni and Marcellino (2014), which applies a dynamic stochastic general equilibrium model (DSGE) to show that potential biases from temporal aggregation can be large in empirical models. Using the Global Trade Analysis Project (GTAP) modeling systems, Brockmeier and Bektasoglu (2014) analyze and compare the effects that data aggregation and model structure have on results. Brockmeier and Bektasoglu (2014) apply both general equilibrium (GE) and partial equilibrium (PE) versions of GTAP, as well as aggregated and disaggregated versions of the input data.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Using the Global Trade Analysis Project (GTAP) modeling systems, Brockmeier and Bektasoglu (2014) analyze and compare the effects that data aggregation and model structure have on results. Brockmeier and Bektasoglu (2014) apply both general equilibrium (GE) and partial equilibrium (PE) versions of GTAP, as well as aggregated and disaggregated versions of the input data. Results show that data aggregation, especially related to competition and tariffs, has a larger effect on model outcome than model structure.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Potential Sources of Aggregation Bias with a Structural Optimization Model of the U.S. Forest Sector

Wade

Baker

Latta

et al. 2019

JfE

View full text Add to dashboard Cite

Structural economic optimization models of the forestry and land use sectors can be used to develop baseline projections of future forest carbon stocks and annual fluxes, which inform policy dialog and investment in programs that maintain or enhance forest carbon stocks. Such analyses vary in terms of the degree of spatial, temporal, and activity-level aggregation used to represent forest resources, land cover, and markets. While the statistical and econometric modeling communities widely discuss the effects of aggregation bias and have developed correction techniques, there is limited prior research investigating how aggregation bias may affect structural optimization models. This paper explores potential aggregation bias using the Land Use and Resource Allocation model (LURA), a detailed spatial allocation partial equilibrium model of the U.S. forest sector. We ran a series of projections representing alternative aggregation approaches including averaging forest stocks at plot, county, state, and regional levels, across one-, five, or ten-year age classes, and by two or fourteen forest types. We compared the resulting projections of forest carbon stocks and harvesting activities across each aggregation scenario. This allows us to isolate the effect of aggregation on key variables of interest (e.g., GHG emissions and supply costs), while holding all other structural characteristics of the modeling framework constant. We find that age-class and forest type aggregations have the greatest impact on modeling results, with the potential to substantially impact market and greenhouse gas projections. On the other hand, spatial aggregation has a small impact on national carbon stock projections. Importantly, regional results are greatly impacted by different aggregation approaches, with projected regional cumulative carbon stocks differing by more than 25% across scenarios.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Evaluating Potential Sources of Aggregation Bias with a Structural Optimization Model of the U.S. Forest Sector

Wade

Baker

Latta

et al. 2019

JfE

View full text Add to dashboard Cite

show abstract

“…The predominant assumption of aggregation bias in the CGE analysis is that a higher degree of sector disaggregation results in larger trade and welfare effects in the simulations performed with CGE models (e.g. Grant et al, 2007Grant et al, , 2008Charteris and Winchester, 2010;Narayanan et al, 2010aNarayanan et al, , 2010bBrockmeier and Bektasoglu, 2014). However, in previous studies, NTBs are not considered, and the AVEs of NTBs that are calculated at different aggregation levels are not compared.…”

Section: Simulation Results: Welfare and Trade Effectsmentioning

confidence: 99%

“…using different data aggregation levels in CGE models and its related effect on simulation results are also acknowledged by several authors (e.g. Grant et al, 2007Grant et al, , 2008Charteris and Winchester, 2010;Alexeeva-Talebi et al, 2012;Brockmeier and Bektasoglu, 2014). Considering NTBs for the policy analysis of RTAs are predominantly done by conducting two-stage analysis; first, the effects of NTBs are estimated using the gravity approach, and then, they are implemented in CGE or partial equilibrium models (e.g.…”

mentioning

confidence: 99%

The Effect of Aggregation Bias: AnNTB‐modelling Analysis of Turkey's Agro‐food Trade with theEU

2016

Self Cite

View full text Add to dashboard Cite

We explore how different data aggregation levels affect the gravity estimates of non‐tariff barriers (NTBs) in the agro‐food sector, and we examine their related impacts on policy simulations of an expansion to the European Union (EU) that would include Turkey. We calculate two sets of ad valorem equivalents (AVEs) of NTBs using the gravity approach to disaggregated and aggregated Central Product Classification data for 15 Global Trade Analysis Project (GTAP) agro‐food sectors. We find that the AVEs of NTBs vary substantially across products and that using aggregated data primarily results in greater effects of NTBs. In a second step, we incorporate the AVEs of NTBs into the GTAP model to evaluate Turkey's EU membership and conclude that aggregation bias has considerable effects on both the estimation of NTBs and the general equilibrium simulation results. Utilising different data aggregation levels leads to a great variability of trade costs of NTBs and, hence, to misleading trade and welfare effects.

show abstract

“…Due to this aggregation presumption, the estimated economic impacts for distinctive sectors simulated at different levels of sector disaggregation might be different, where a low degree could produce distorted assessment results (Alexeeva-Talebi et al, 2012;Caron, 2012;Brockmeier & Bektasoglu, 2014). As a result of the diverse selection of IO-tables in different resolution levels of sector aggregation, the selected level needs to be justifiable to answer research questions or provide the economic policy analysis appropriately instead of an ad hoc selection, in order to refrain from inappropriate models that violated substitutability assumption or inaccurate values of elasticities (Shumway & Davis, 2001).…”

Section: Introductionmentioning

confidence: 99%

The Impact of Sectoral Aggregation on Elasticities of Substitution Using Translog Cost Function

Hardadi¹,

Pauliuk²

2021

Preprint

View full text Add to dashboard Cite

Economic models are widely used to simulate policy scenarios, in which elasticities are used as measures of substitutability between production inputs. These models need appropriate levels of sector aggregation to produce policy-relevant results and to capture the variations of input substitutability among different sectors, while still adhering to the monotonicity and concavity requirements of aggregate production function. The purposes of this study are (i) to assess the cost (production) functions that fulfill these requirements appropriately, (ii) to analyze the impact of sector aggregation on the elasticities, and (iii) to determine the optimum resolution level for modelling cost function. This study utilized two databases to construct an industry-level panel dataset for 1995–2016, EU-KLEMS to obtain the price indices of the capital and labor inputs, and time-series monetary EXIOBASE v3.6, in both current and constant prices, to obtain the monetary inputs and price indices of intermediate inputs. Dynamic translog in GMM estimation is selected to derive the elasticities of substitution on different levels of sector aggregation since it performs better in reducing concavity violations, compared to pooled and fixed-effect estimation method. Selecting a higher resolution level is preferred to produce better model fittings, but it occasionally results in more concavity violations. Modelling cost functions at a more aggregated level leads to a larger estimate of the elasticity of substitution. The optimized level of sector aggregation for modelling cost function obtained in this study is at 86 industry sectors, capturing the different elasticities of substitution in various mining and manufacturing sectors of basic materials and electricity production.

show abstract

Model structure or data aggregation level: Which leads to greater bias of results?

Cited by 15 publications

References 11 publications

Evaluating Potential Sources of Aggregation Bias with a Structural Optimization Model of the U.S. Forest Sector

Evaluating Potential Sources of Aggregation Bias with a Structural Optimization Model of the U.S. Forest Sector

The Effect of Aggregation Bias: AnNTB‐modelling Analysis of Turkey's Agro‐food Trade with theEU

The Impact of Sectoral Aggregation on Elasticities of Substitution Using Translog Cost Function

Contact Info

Product

Resources

About