Dissertations / Theses on the topic 'Production function estimation'

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Economics, May, 2020
Cataloged from the official PDF of thesis.
Includes bibliographical references (pages 193-201).
This first chapter develops a new method for estimating production functions with factor-augmenting technology and assesses its economic implications. The method does not impose parametric restrictions and generalizes prior approaches that rely on the CES production function. I first extend the canonical Olley-Pakes framework to accommodate factor-augmenting technology. Then, I show how to identify output elasticities based on a novel control variable approach and the optimality of input expenditures. I use this method to estimate output elasticities and markups in manufacturing industries in the US and four developing countries. Neglecting labor-augmenting productivity and imposing parametric restrictions mismeasures output elasticities and heterogeneity in the production function. My estimates suggest that standard models (i) underestimate capital elasticity by up to 70 percent (ii) overestimate labor elasticity by up to 80 percent.
These biases propagate into markup estimates inferred from output elasticities: markups are overestimated by 20 percentage points. Finally, heterogeneity in output elasticities also affects estimated trends in markups: my estimates point to a much more muted markup growth (about half) in the US manufacturing sector than recent estimates. The second chapter develops partial identification results that are robust to deviations from the commonly used control function approach assumptions and measurement errors in inputs. In particular, the model (i) allows for multi-dimensional unobserved heterogeneity,(ii) relaxes strict monotonicity to weak monotonicity, (iii) accommodates a more flexible timing assumption for capital. I show that under these assumptions production function parameters are partially identified by an 'imperfect proxy' variable via moment inequalities. Using these moment inequalities, I derive bounds on the parameters and propose an estimator.
An empirical application is presented to quantify the informativeness of the identified set. The third chapter develops an approach in which endogenous networks is a source of identification in estimations with network data. In particular, I study a linear model where network data can be used to control for unobserved heterogeneity and partially identify the parameters of the linear model. My method does not rely on a parametric model of network formation. Instead, identification is achieved by assuming that the network satisfies latent homophily - the tendency of individuals to be linked with others who are similar to themselves. I first provide two definitions of homophily: weak and strong homophily. Then, based on these definitions, I characterize the identified sets and show that they are bounded under weak conditions.
Finally, to illustrate the method in an empirical setting, I estimate the effects of education on risk preferences and peer effects using social network data from 150 Chinese villages.
by Mert Demirer.
Ph. D.
Ph.D. Massachusetts Institute of Technology, Department of Economics

Robustness of DEA scale efficiency scores is investigared in the context of non-radial efficency measures. Most efficient scales are identified with DEA's refernce firms instead of traditional clustering techniques.The systematic difference between single- and multi-output technologies as concerns most efficient scales is then examined by comparing applied DEA results. These provide evidence in favor of my proposition, that single-output techniques, like stochastic production function estimation, yield upwardly biased most efficient scales.
Series: Department of Economics Working Paper Series

Mestrado em Econometria Aplicada e Previsão
Vários estudos econométricos procuram explicar os determinantes da criação de conhecimento usando como variável dependente o número de patenteamentos numa região. Alguns destes procuram captar os efeitos de Knowledge Spillovers através de modelos lineares que incorporam dependência espacial. No entanto, nenhum estudo foi encontrado que captasse este efeito, tendo em atenção a natureza discreta da variável dependente. Este trabalho pretende preencher essa lacuna propondo um novo estimador de máxima verosimilhança a dois passos para um modelo Poisson Autorregressivo Espacial. As propriedades do estimador são avaliadas num conjunto de simulações de Monte Carlo. Os resultados sugerem que este estimador tem menor Bias e menor RMSE, na generalidade, que outros estimadores propostos, sendo que apenas mostra piores resultados quando a dependência espacial é próxima da unidade. Um exemplo empírico, empregando o novo estimador e um conjunto de estimadores alternativos, é realizado, sendo que a criação de conhecimento em 234 NUTS II de 24 países europeus é analisada. Os resultados evidenciam que existe uma forte dependência espacial na criação de inovação entre as regiões. Conclui-se também que o ambiente socioeconómico é essencial para o processo de formação de conhecimento e que contrariamente às instituições públicas, as empresas privadas são eficientes na produção de inovação. É de realçar, que regiões com menor capacidade em transformar despesas R&D em patenteamentos apresentam maior capacidade de absorção e segregação de conhecimento, evidenciando que regiões vizinhas menos eficientes na produção de conhecimento tendem a criar relações fortalecidas na partilha de conhecimento.
Several econometric studies seek to explain the determinants of knowledge production using as dependent variable the number of patents in a region. Some of these capture the effects of knowledge spillovers through linear models with spatial autorregressive term. However, no study has been found that estimates such effect while also considering the discrete nature of the dependent variable: a count variable. This essay aims to fill this gap by proposing a new Two-step Maximum Likelihood estimator for a Spatial Autorregressive Poisson model. The properties of this estimator are evaluated in a set of Monte Carlo Experiments. The simulation results suggest that this estimator presents lower Bias and lower RMSE than the alternative estimators proposed, only showing worse results when the spatial dependence is close to the unit. An empirical example, using the new estimator and a set of alternative estimators, is executed, where the creation of knowledge in 234 NUTS II from 24 European countries is analyzed. The results show that there is a strong spatial dependence on the creation of innovation. It is also concluded that the socio-economic environment is essential for the knowledge formation and, unlike public R&D institutions, private companies are efficient in producing innovation. It should be noted that regions with less capacity to transform R&D expenses into new patents, have greater capacity for absorption and segregation of knowledge, which shows that neighboring regions less efficient in the production of knowledge tend to create strong relations with each other taking advantage of the knowledge sharing process.
info:eu-repo/semantics/publishedVersion

La fonction de production des entreprises lie leur niveau de production à leurs dépenses en facteurs de production. Son estimation est à la fois importante et peu fiable. Importante, parce que des indicateurs clés pour la conception des politiques publiques, tels que le taux de marge, en découlent. Peu fiable, car elle repose sur des hypothèses d'identification.Ce projet étudie les hypothèses qui sous-tendent l'estimation des fonctions de production : à la fois leurs formes fonctionnelles et la flexibilité des facteurs de production. Il s'appuie ensuite sur une estimations des fonctions de production des entreprises françaises, pour évaluer l'évolution de leur taux de marge et de la part du travail dans leur valeur ajoutée au cours des 30 dernières années.Le chapitre 1 conforte l'hypothèse de flexibilité des facteurs de production: l'ajustement instantané des matières premières ou du travail et l'ajustement retardé du capital. Nous nous appuyons sur l'existence de notches ; des valeurs où les bénéfices après impôt diminuent avec le chiffre d’affaire avant impôt dans le code des impôts français.Nous montrons que les entreprises qui optimisent ont une plus grande élasticité de production par rapport aux matières premières et une plus faible élasticité de la production par rapport au capital. De même, pour ajuster leur production, les entreprises ont tendance à réduire principalement leurs dépenses en matières premières.Le chapitre 2 s’appuie sur le résultat du chapitre 1 pour mesurer le taux de marge de toutes les entreprises françaises entre 1984-2016. De Loecker et Warzynski (2012) montrent que la marge d'une entreprise est proportionnelle à l'inverse de la part de revenu de l'un de ses intrants flexibles. Nous analysons l'évolution des marges agrégées en France et documentons que l'augmentation de la concentration est corrélée à une réallocation des parts de marché vers les entreprises à marge élevée. Nous montrons également que l'évolution de la part du travail reflète l'évolution des marges : la réallocation tend à diminuer la part du travail tandis qu'au sein des entreprises, la part du travail augmente.Le choix d'une forme fonctionnelle pour décrire le processus de production est un compromis entre théorie et empirisme. La fonction de production standard est de type Cobb-Douglas mais impose une élasticité de substitution constante et égale à 1, en contradiction avec la littérature empirique. Les fonctions de production CES ont des élasticités de substitution non unitaires mais constantes au sein de chaque industrie et un ratio d'utilisation des facteurs de production indépendant de la taille de l'entreprise.Le chapitre 3 montre que cette dernière fonction de production ne rend pas compte de l'utilisation des technologies de l’information (TIC), puisque nous documentons une augmentation de la demande relative de TIC avec la taille des entreprises: en cohérence avec une fonction de production CES non-homothétique. Nous analysons ensuite comment l'interaction de la baisse des prix des TIC et les caractéristiques non-homothétiques des TIC rationalisent les faits empiriques documentés dans le chapitre 2. (i) comme les grandes entreprises sont plus intensives en TIC, elles bénéficient de manière disproportionnée de la baisse des prix des TIC, ce qui rationalise l'augmentation de la concentration. (ii) comme les grandes entreprises sont plus intensives en TIC dans l'échantillon, elles fonctionnent avec des rendements d'échelle plus faibles et ont donc des parts de bénéfices plus élevées et des parts de travail plus faibles. Cela explique comment l'augmentation de la concentration entraîne une diminution de la part globale du travail. (iii) les statistiques comparatives du modèle prédisent que l’adoption de TIC liée à la baisse de leur prix implique des rendements d'échelle plus élevés et ont donc une part de travail plus importante, ce qui explique la tendance haussière de la part du travail au sein des entreprises
Firms production function link their use of input factors to their production level. Production function estimates are at the same time important and untrustworthy. Important, because key indicators for policy design, such as the measure of aggregate markups, are derived from those estimates. Untrustworthy because they rely on identification assumptions.This project studies the assumptions underlying the usual techniques for estimating production functions: both their functional forms and the often assumed inputs flexibility. It then leverages production function estimates, to assess how firms ability to price over marginal income and the share of their income going to workers have evolved over the last 30 years.In Chapter 1 we provide evidence on the input flexibility assumption grounding production function estimation: the quasi instantaneous adjustment of either material or labor and delayed adjustment of capital hold. We rely on the existence of notches; values where after-tax profits decrease in before-tax sales in the French tax code.We identify which type of firms adjust their size in response to a transient notch. We do this by studying the ex-ante characteristics of firms below the tax cutoff. We find that firms who bunch tend to have larger elasticity of output with respect to materials and lower elasticity of output with respect to capital. Consistently, we also show that to adjust their remaining production firms, tend to primarily reduce spending on material.In Chapter 2 we leverage evidence on inputs flexibility to recover firm level markups of the universe of firms in France over the 1984-2016 period. De Loecker and Warzynski (2012) show that a firm’s markup proportionates the inverse of one of its flexible inputs revenue share. We analyze the evolution of aggregate markups in France and document that the rise of concentration correlates with a reallocation of market share towards high markup firms. We also show that the evolution of the labor share mirrors the evolution of markups: reallocation tends to decrease labor share while within firms, labor share rises.The choice of a functional form to describe firms production process is a compromise between theory and empirics. The workhorse production function is Cobb-Douglas and imposes a constant (and equal to 1) elasticity of substitution. Recent evidence in the empirical literature has however estimated a micro-elasticity of substitution significantly lower than one. While CES production functions allow for non-unit elasticity of substitution, they assume constant elasticity within industry and imply that the ratio of input use doesn’t depend on firm size.In Chapter 3, we show that the latter production function cannot account for IT inputs use in firms.With detailed data on software and hardware investments among French firms, we document that the firm-level demand for IT inputs relative to other inputs grows in the firm’s scale of operation. Theoretically, a non-homothetic CES production function helps rationalizing this empirical fact.We then analyze how the interaction of the fall of IT prices and the non-homothetic characteristics of IT inputs also help rationalize the empirical facts documented in chapter 2. First, since larger firms are more IT intensive in the cross-section, they benefit disproportionally from the fall in IT prices, rationalizing the rise of concentration. Similarly, since larger firms are more IT intensive in the cross-section, they operate at lower returns to scale and therefore have higher profit shares and lower labor shares. This explains how the rise in concentration drives a decline in aggregate labor shares. Finally, the comparative statistics of the model predicts that the fall of IT prices imply that when firms substitute toward IT they operate at higher returns to scale and therefore tend to have larger labor share, explaining the positive contribution to aggregate labor share of the within component

Les simulations d'écoulement sur des modèles représentatifs d'un gisement pétrolier sont généralement coûteuses en temps de calcul. Une pratique courante en ingénierie de réservoir consiste à remplacer ces simulations par une approximation mathématique, un méta-modèle. La méta-modélisation peut fortement réduire le nombre de simulations nécessaires à l'analyse de sensibilité, le calibrage du modèle, l'estimation de la production, puis son optimisation. Cette thèse porte sur l'étude de méta-modèles utilisant des simulations réalisées à différents niveaux de précision, par exemple pour des modèles de réservoir avec des maillages de résolutions différentes. L'objectif est d'accélérer la construction d'un méta-modèle prédictif en combinant des simulations coûteuses avec des simulations rapides mais moins précises. Ces méta-modèles multi-fidélité, basés sur le co-krigeage, sont comparés au krigeage pour l'approximation de sorties de la simulation d'écoulement. Une analyse en composantes principales peut être considérée afin de réduire le nombre de modèles de krigeage pour la méta-modélisation de réponses dynamiques et de cartes de propriétés. Cette méthode peut aussi être utilisée pour améliorer la méta-modélisation de la fonction objectif dans le cadre du calage d'historique. Des algorithmes de planification séquentielle d'expériences sont finalement proposés pour accélérer la méta-modélisation et tirer profit d'une approche multi-fidélité. Les différentes méthodes introduites sont testées sur deux cas synthétiques inspirés des benchmarks PUNQ-S3 et Brugge
Performing flow simulations on numerical models representative of oil deposits is usually a time consuming task in reservoir engineering. The substitution of a meta-model, a mathematical approximation, for the flow simulator is thus a common practice to reduce the number of calls to the flow simulator. It permits to consider applications such as sensitivity analysis, history-matching, production estimation and optimization. This thesis is about the study of meta-models able to integrate simulations performed at different levels of accuracy, for instance on reservoir models with various grid resolutions. The goal is to speed up the building of a predictive meta-model by balancing few expensive but accurate simulations, with numerous cheap but approximated ones. Multi-fidelity meta-models, based on co-kriging, are thus compared to kriging meta-models for approximating different flow simulation outputs. To deal with vectorial outputs without building a meta-model for each component of the vector, the outputs can be split on a reduced basis using principal component analysis. Only a few meta-models are then needed to approximate the main coefficients in the new basis. An extension of this approach to the multi-fidelity context is proposed. In addition, it can provide an efficient meta-modelling of the objective function when used to approximate each production response involved in the objective function definition. The proposed methods are tested on two synthetic cases derived from the PUNQ-S3 and Brugge benchmark cases. Finally, sequential design algorithms are introduced to speed-up the meta-modeling process and exploit the multi-fidelity approach

Les simulations d'écoulement sur des modèles représentatifs d'un gisement pétrolier sont généralement coûteuses en temps de calcul. Une pratique courante en ingénierie de réservoir consiste à remplacer ces simulations par une approximation mathématique, un méta-modèle. La méta-modélisation peut fortement réduire le nombre de simulations nécessaires à l'analyse de sensibilité, le calibrage du modèle, l'estimation de la production, puis son optimisation. Cette thèse porte sur l'étude de méta-modèles utilisant des simulations réalisées à différents niveaux de précision, par exemple pour des modèles de réservoir avec des maillages de résolutions différentes. L'objectif est d'accélérer la construction d'un méta-modèle prédictif en combinant des simulations coûteuses avec des simulations rapides mais moins précises. Ces méta-modèles multi-fidélité, basés sur le co-krigeage, sont comparés au krigeage pour l'approximation de sorties de la simulation d'écoulement. Une analyse en composantes principales peut être considérée afin de réduire le nombre de modèles de krigeage pour la méta-modélisation de réponses dynamiques et de cartes de propriétés. Cette méthode peut aussi être utilisée pour améliorer la méta-modélisation de la fonction objectif dans le cadre du calage d'historique. Des algorithmes de planification séquentielle d'expériences sont finalement proposés pour accélérer la méta-modélisation et tirer profit d'une approche multi-fidélité. Les différentes méthodes introduites sont testées sur deux cas synthétiques inspirés des benchmarks PUNQ-S3 et Brugge
Performing flow simulations on numerical models representative of oil deposits is usually a time consuming task in reservoir engineering. The substitution of a meta-model, a mathematical approximation, for the flow simulator is thus a common practice to reduce the number of calls to the flow simulator. It permits to consider applications such as sensitivity analysis, history-matching, production estimation and optimization. This thesis is about the study of meta-models able to integrate simulations performed at different levels of accuracy, for instance on reservoir models with various grid resolutions. The goal is to speed up the building of a predictive meta-model by balancing few expensive but accurate simulations, with numerous cheap but approximated ones. Multi-fidelity meta-models, based on co-kriging, are thus compared to kriging meta-models for approximating different flow simulation outputs. To deal with vectorial outputs without building a meta-model for each component of the vector, the outputs can be split on a reduced basis using principal component analysis. Only a few meta-models are then needed to approximate the main coefficients in the new basis. An extension of this approach to the multi-fidelity context is proposed. In addition, it can provide an efficient meta-modelling of the objective function when used to approximate each production response involved in the objective function definition. The proposed methods are tested on two synthetic cases derived from the PUNQ-S3 and Brugge benchmark cases. Finally, sequential design algorithms are introduced to speed-up the meta-modeling process and exploit the multi-fidelity approach

Made available in DSpace on 2015-03-26T13:50:06Z (GMT). No. of bitstreams: 1 texto completo.pdf: 1390780 bytes, checksum: 772c85b924a2215ba464b70773a148bd (MD5) Previous issue date: 2010-10-28
The sugarcane crop is subjected during their growth to different environmental conditions, and agricultural yield directly affected by these conditions. Yield models become important tools in order to supply estimates of crop yield along to characterize management alternatives, increasing the efficiency of management and strategic decisions. Information technology is an important tool in this process and has been increasingly used for collecting and analyzing data that are used as the basis of their decisions. The objective of this work was included in the software Irriplus estimated yield of crops models, using the Stewart and Mantovani models to estimate the real productivity and Method of Agroecological Zone (MAZ) to estimate the maximum productivity. Besides the models, a methodology of multiple linear regressions was developed to explain the factors that are influencing the yield crop and generate yield models from historical data. To evaluate the models, was used descriptive analysis and analysis tests comparing the real and estimated yield. Statistical tests were paired t-test, relative error percentage (REP) and mean absolute error (MAE). Was used real yield data of irrigated sugarcane RB 86-7515, crops xiv 2007/2008 and 2008/2009, in the Jaíba city of Minas Gerais state. The Stewart Model requires as input the maximum yield that was estimated by MAZ in two crops. In the 2007/2008 crop, the model estimated the average productivity at 113.58 t ha-1, while the real yield average was 113.47 t ha-1, the MAE was equal 10.10. In the 2008/2009 crop, the model estimated the average productivity at 121.81 t ha-1, while the real yield average was 121.81 t ha-1, the MAE was equal 8.02. In both crops the paired t-test showed no significant difference between the average yields. The Mantovani model used the same maximum yield of the model of Stewart, estimated by MAZ. In the 2007/2008 crop, the model estimated the average productivity at 198.13 t ha-1, while the real yield average was 113.47 t ha-1, the MAE was equal 84.66. In the 2008/2009 crop, the model estimated the average productivity at 154.81 t ha-1, while the real yield average was 121.81 t ha-1, the MAE was equal 32.72. In both crops the paired t-test showed significant difference between the average yields and the yield estimated overestimated the real yield. An equation was fitted by multiple linear regression using data from the 2007/2008 crop, related variables: total irrigation required, total capacity of the soil water, available soil water, reference evapotranspiration, crop evapotranspiration and maximum crop evapotranspiration. The equation was evaluated in the 2008/2009 crop for yield estimated. The equation estimated the average productivity at 122.41 t ha-1, while the real yield average was 121.81 t ha-1, the MAE was equal 7.07. The paired t-test showed no significant difference between the average yields.
A cultura da cana-de-açúcar é submetida durante o seu desenvolvimento a diferentes condições ambientais, sendo o rendimento agrícola afetado diretamente por estas condições. Modelos de produtividade tornam-se ferramentas importantes objetivando suprir estimativas de rendimento ao longo das safras visando à caracterização de alternativas de manejo, aumentando a eficácia das decisões gerenciais e estratégicas. A tecnologia da informação é uma importante ferramenta nesse processo e tem sido cada vez mais utilizada para coleta e análise de dados que são utilizados como base nas suas decisões. O objetivo deste trabalho foi incluir no software Irriplus modelos de estimativa de produtividades de culturas agrícolas, utilizando os modelos de Stewart e Mantovani para estimar a produtividade real e o Método da Zona Agroecológica (MZA) para estimar a produtividade máxima. Além dos modelos, foi desenvolvida uma metodologia de regressão linear múltipla para explicar os fatores que estão influenciando a produtividade da cultura e gerar modelos de produtividade a partir de dados históricos. Para avaliar os modelos, foi utilizada análise descritiva e testes de análise comparativa entre a produtividade estimada e observada em campo. Os testes estatísticos utilizados foram: teste-t pareado, erro relativo percentual (ERP) e erro médio absoluto (MAE). Foram utilizados dados reais de produtividade da cana-de-açúcar RB 86-7515 irrigada, safras 2007/2008 e 2008/2009, do município de Jaíba do estado de Minas Gerais. O modelo de Stewart requer como dado de entrada a produtividade máxima, que foi estimada pelo MZA nas duas safras. Na safra 2007/2008, o modelo estimou a produtividade média em 113,58 t ha-1, enquanto a produtividade média observada em campo foi 113,47 t ha-1, o MAE foi igual a 10,10. Na safra 2008/2009 o modelo estimou a produtividade média em 121,81 t ha-1, enquanto a produtividade média observada em campo foi 121,81 t ha-1, o MAE foi igual a 8,02. Nas duas safras o teste-t pareado não demonstrou diferença significativa entre as médias de produtividade. O modelo de Mantovani utilizou a mesma produtividade máxima do modelo de Stewart estimada pelo MZA. Na safra 2007/2008, o modelo estimou a produtividade média em 198,13 t ha-1, enquanto a produtividade média observada em campo foi 113,47 t ha-1, o MAE foi igual a 84,66. Na safra 2008/2009, o modelo estimou a produtividade média em 154,81 t ha-1, enquanto a produtividade média observada em campo foi 121,81 t ha-1, o MAE foi igual a 32,72. Nas duas safras, o teste-t pareado demonstrou diferença significativa entre as médias de produtividade e a estimativa do modelo superestimou produtividade observada em campo. Foi ajustada uma equação por regressão linear múltipla, com dados da safra 2007/2008, relacionada com as variáveis: irrigação total necessária, capacidade total de água no solo, água disponível no solo, evapotranspiração de referência, evapotranspiração da cultura e evapotranspiração máxima da cultura. A equação foi avaliada na safra 2008/2009 para estimativa da produtividade. A equação estimou a produtividade média em 122,41 t ha-1, enquanto a produtividade média observada em campo foi 121,81 t ha-1, o MAE foi igual a 7,07. O teste-t pareado não demonstrou diferença significativa entre as médias de produtividade.

The purpose of this thesis is to investigate the levels of interdependency (simultaneity) between the performances of two direct opponents in professional football. More specifically, interdependency between performances is investigated at three different levels. Firstly, empirical analyses are conducted in order to assess the levels of interdependency between the performances of two teams in direct competition using team-game observations. Secondly, data on team formations and player starting positions within these formations is utilised in order to uniquely match individual players to a sole opponent on the field of play. Further empirical analyses are then conducted in order to investigate the levels of interdependency at this more isolated individual player level using player-game observations. Finally, an empirical investigation into the levels of interdependency between the performances of a positional unit (defence or attack) and their opposing team (as a collective) is conducted using positional unit-game observations. An exclusive and detailed data set ranging from the 2007-08 season to the 2010-11 season is utilised in order to estimate several production functions for teams, individual players and positional units competing in the German Bundesliga. The results in all empirical analyses confirm that the performance of the opponent is significant. At a team and positional unit level, no evidence is found to suggest that the performances of two direct opponents(s) are interdependent, however the results reveal that the recent past performances of the opponent(s) have a significant linear impact upon the performance of the subject. In particular, relative team form going in to a match is revealed to have a significant impact upon the performance of their opponent. At an individual player level, evidence is found to confirm that the performances of players in direct competition are interdependent thus supporting the sports economics theory of joint production. Specifically, the results reveal that the performances of defenders have a significant and negative impact upon the performances of their opposing attackers.

Submitted by Salvio Rocha (salviorocha@gmail.com) on 2013-09-27T17:07:32Z No. of bitstreams: 1 Salvio-MFEE-FinalCompleto.pdf: 1969810 bytes, checksum: 8246730df4d70a0a3e57ce52fbe79b5e (MD5)
Approved for entry into archive by Vitor Souza (vitor.souza@fgv.br) on 2013-10-02T14:32:15Z (GMT) No. of bitstreams: 1 Salvio-MFEE-FinalCompleto.pdf: 1969810 bytes, checksum: 8246730df4d70a0a3e57ce52fbe79b5e (MD5)
Made available in DSpace on 2013-10-03T17:02:40Z (GMT). No. of bitstreams: 1 Salvio-MFEE-FinalCompleto.pdf: 1969810 bytes, checksum: 8246730df4d70a0a3e57ce52fbe79b5e (MD5) Previous issue date: 2013-05-24
Productivity is often calculated by the approach of the Cobb-Douglas production function. This estimate, however, may suffer from s imultaneity and inputs bias selection. The method of Olley and Pakes (1996) int roduced a semi-parametric method allows us to estimate the parameters of the production function consistently and thus obtain reliable measures of p roductivity, controlling such bias problems. This study applies this method in a company the sugarcane sector and uses the command opreg at Stata to esti mate the production function of a company and concluded with the econom ic intuition behind the result.
Produtividade é frequentemente calculada pela aproximação da função de produção Cobb-Douglas. Tal estimativa, no entanto, pode sofrer de simultaneidade e viés de seleção dos insumos. Olley e Pakes (1996) introduziu um método semi-paramétrico que nos permite estimar os parâmetros da função de produção de forma consistente e, assim, obter medidas de produtividade confiável, controlando tais problemas de viés. Este estudo aplica este método em uma empresa do setor sucroalcooleiro e utiliza o comando opreg do Stata com a finalidade de estimar a função produção, descrevendo a intuição econômica por trás dos resultados.

The PhD thesis developed an economic model as an integral part of the current Health Impact Assessment (HIA) framework. Based on a Health Production Function approach, the model showed how to estimate economic benefits of positive health gains generated by transport investment programs and transport policies. Using Australian mortality and morbidity statistics and applying econometric analysis, the case study quantified health benefits induced by transport emission abatement policies in dollar terms for the Australian households. Finally, the thesis demonstrated transferability of the economic model through two example case studies, establishing a wider application capacity of the model.

This study estimates the economic costs of mining pollution on water resources for the years 2008 and 2009 based on the conceptual framework of Environmental Efficiency. This framework identifies such costs as the mining companies’ trade-off between increasing production that is saleable at market prices (desirable output) and reducing the environmental pollution that emerges from the production process (undesirable output). These economic costs were calculated from parametric and non parametric production possibility frontiers for 28 and 37 mining units in 2008 and 2009, respectively, which were under the purview of the National Campaign for Environmental Monitoring of Effluent and Water Resources, conducted by the Energy and Mining Investment Supervisory Agency (Osinergmin) in those years. The results show that the economic cost of mining pollution on water resources rose to U.S. $ 814.7 million and U.S. $ 448.8 million for 2008 and 2009, respectively. These economic costs were highly concentrated in a few mining units, within a few pollution parameters, and were also higher in mining units with average/low mineral production. Taking into consideration that at present the fine and penalty system in the mining sector is based on administrative criteria, this study proposes a System of Environmentally Efficient Sanctions based on economic criteria so as to establish a preventive mechanism for pollution. It is hoped that this mechanism will generate the necessary incentives for mining companies to address the negative externalities that emerge from their production process.
En este estudio se aproximan los costos económicos de la contaminación ambiental minera sobre los recursos hídricos para 2008 y 2009 en el marco conceptual de la Eficiencia Medioambiental, que interpreta dichos costos como el trade-off de los empresarios mineros entre incrementar su producción que es vendible a precios de mercado (output deseable) yreducir la contaminación ambiental que se desprende de su proceso productivo (output no deseable). Dichos costos económicos fueron calculados a partir de fronteras de posibilidades de producción paramétricas y no paramétricas para 28 y 37 unidades mineras en los años 2008 y 2009 respectivamente, las que estuvieron bajo el ámbito de la Campaña Nacional deMonitoreo Ambiental de Efluentes y Recursos Hídricos que realizó el Organismo Supervisor de Inversión Energía y Minería (Osinergmin) en dichos años. Los resultados indican que los costos económicos de la contaminación ambiental minera sobre los recursos hídricos ascendieron, en promedio, para los años 2008 y 2009, a US$ 814,7 millones,y US$ 448,8 millones, respectivamente. Dichos costos estuvieron altamente concentrados en pocas unidades productivas, así como en pocos parámetros de contaminación, y fueron mayores en unidades mineras con producción media/baja de minerales. Dado que en la actualidad el sistema de multas y sanciones en el sector minero se basa en criterios administrativos, el estudio propone un Sistema de Sanciones Ambientalmente Eficiente basado en criterios económicos

Made available in DSpace on 2010-04-20T20:58:08Z (GMT). No. of bitstreams: 4 JOSE_ANTONIO_DE_SOUZA_DISSERTACAO_VERSAO_FINAL.pdf.jpg: 17403 bytes, checksum: e10358d7e12871cef2b0316788b3e376 (MD5) JOSE_ANTONIO_DE_SOUZA_DISSERTACAO_VERSAO_FINAL.pdf.txt: 192545 bytes, checksum: bd219ee8e4077c7534538b47349df885 (MD5) license.txt: 4810 bytes, checksum: 5a0053361f1e90aa2d2231437718e9df (MD5) JOSE_ANTONIO_DE_SOUZA_DISSERTACAO_VERSAO_FINAL.pdf: 761735 bytes, checksum: 652c91e503be04b555cfabf4dcbabf1e (MD5) Previous issue date: 2009-11-26T00:00:00Z
The basic hypothesis and core of this dissertation is that various methods of estimating production functions produce different results when applied to sectors of different technological intensity. This dissertation focused on determining the total factor productivity in several industries. Four sectors with high technological intensity and four sectors of low technology intensity were selected for assess this hypothesis. Production functions were estimated and its residue used to calculate the productivity. The correlation between residuals and the explanatory variables inherent to this procedure, including that of simultaneity, omitted variables and selection, was taken into account. One goal of this study was to identify whether a particular method would be more suitable to estimate production functions for industries with low/high technological intensity. This work studied several methods to estimate production functions, including: Olley & Pakes, and Levinsohn & Petrin. Our results show that, for industries with low and with high technological intensity, the Olley & Pakes method estimates are marginally better than the ones from Levinsohn & Petrin. In our opinion, such results do not provide enough advantage to put away the Levinsohn & Petrin method as a method to estimate production functions. The sensitivity of results to the different methods suggests that all of them should be consulted. In addition to the previously stated results, this work identified that the sectors studied experienced a productivity decline or stagnation from 1996 to 2005.
Este trabalho investigou o problema da determinação da produtividade total dos fatores em diversos setores industriais. Tal determinação se dá por meio de estimação de funções de produção, obtendo-se a produtividade a partir do resíduo destas estimações. A questão que aflora deste procedimento é a existência de correlação entre os resíduos e as variáveis explicativas, implicando em diversos vieses, entre eles o de simultaneidade, de variáveis omitidas e de seleção. Neste trabalho foram abordados diversos métodos de estimação de funções de produção, entre eles os métodos de Olley e Pakes e Levinsohn e Petrin. Todos os métodos foram aplicados a diversos setores da economia. A escolha dos setores se deu com base na intensidade tecnológica de cada um, sendo então escolhidos quatro setores de alta intensidade tecnológica e quatro de baixa intensidade tecnológica. A hipótese básica, fio condutor deste trabalho, é que os diversos métodos de estimação de funções de produção apresentam diferentes resultados quando aplicados a setores de diferentes intensidades tecnológicas. Um dos objetivos deste estudo foi identificar se determinado método seria mais adequado a setores de baixa intensidade tecnológica, enquanto outro seria mais apropriado a setores de alta intensidade tecnológica. Conclui-se que o método de Olley e Pakes é levemente superior ao de Levinsohn e Petrin em ambos os grupos de setores, mas não a ponto de se descartar o segundo método. A sensibilidade dos resultados aos diferentes métodos sugere que todos devem ser consultados. Um resultado adicional deste trabalho é a constatação de que houve queda ou estagnação da produtividade nos setores selecionados para a década de 1996 a 2005.

What motivates mergers in banking? The data show that merger activity is concentrated among very large banks. A large literature on the banking structure has studied this question by estimating cost functions and has provided mixed evidence. A crucial assumption is the exogeneity of input prices.If this assumption fails, result may be biased. This paper adopts the production function method proposed by Levinsohn and Petrin (2003) to separate the impact of productivity from scale economies in banking. To avoid this bias, I use recovery rates of non-performing loans, charge off rates, and cash holdings as proxies for productivity. The proxy method illustrates that the industry operates with significant diseconomies of scale, while the OLS method generates opposite results. Therefore, this finding supports the view that improvements in productivity cause mergers, which is also consistent with data. Finally, I introduce the Quantile Proxy Method to capture the impacts of both input endogeneity and size heterogeneity. This method reveals that medium size banks have largest diseconomies of scale, while top 5% experience somewhat extensive economies of scale. This result sheds light on the fact why many mergers occur among large banks: large parties involved in a consolidation benefit from both productivity improvements and scale economies.
text

碩士
世新大學
經濟學研究所(含碩專班)
98
The study tackles the dispersion of the regional of specified economic growth in Taiwan. Our results show that regional growth patterns can be understood as a function of several interrelated factors, which include investment in physical capital, human capital, and infrastructure capital; the infusion and spread of new technology. We find that human capital positively affects output per worker and productivity growth. In particular, in terms of its direct contribution to production, educated labor has a much higher marginal product. This direct effect is hypothesized to come from domestic innovation activities. The estimated spillover effect of human capital on TFP growth is positive and statistically significant, which is very robust to model specifications and estimation methods. We conduct a cost-benefit analysis and a policy experiments in which we project the impact of increases in human capital and infrastructure capital on regional inequality. We conclude that investing in human capital will be an effective policy to reduce regional gaps in Taiwan as well as an efficient means to promote economic growth.

Our goal is the estimation of country - level production function aimed at understanding the role of human capital. We analyze the effect of education, especially the effect of the share of college graduates in prime-age population (between 25 - 54 years) on the European Union (EU) countries' labor productivity. Here, an important issue is efficiency of tertiary education institutions. We split the ratio of human capital to observe it from different aspects. We compare an effect of lower and upper tertiary educated, by specializations and by gender. The relationship between human capital and labor productivity was found positive though not significant or significant only on 10 % confidence level. The influence of human capital on labor productivity was found very low, in some cases even negative. Assuming that one of the main reasons behind these contra-intuitive results is the problem of unobserved heterogeneity, we also run instrumental variable estimation. We found positive and significant on 5 % confidence level relationship between human capital and labor productivity. The influence of larger share of tertiary educated people on labor productivity is more evident after some period of time, in our example after two years. Keywords: human capital, labor productivity, European Union, production...

The study here examines the performance of Tasmanian public primary schools over the period of 2000 to 2007. The three principal objectives of the study are (i) to investigate the effects of school resources on students‘ academic achievement; (ii) evaluate schools‘ technical efficiency; and (iii) to identify the factors that affect the schools‘ technical efficiency. To explore the first objective, an educational production function is estimated. The second and the third objectives are examined by employing two techniques, namely, Stochastic Production Frontier (SPF) and Data Envelopment Analysis (DEA). On the basis of the estimation of a Fixed Effects model of a Tasmanian educational production function, a one per cent increase in educational expenditure per student is associated with an increase in the reading, writing and numeracy scores of 0.38%, 0.36% and 0.43%, ceteris paribus. Male students‘ performance was on average lower than female students. Evidence of relatively lower performance by indigenous students was found. Students‘ performance was also negatively affected by their level of absenteeism. Positive effects of parental education on Tasmanian students‘ academic achievement were found but the effects of parental occupation were not statistically significant. The SPF estimates suggest that Tasmanian public primary schools are almost technically efficient with the average technical efficiency score constant at 97% from 2003 to 2007. No technical efficiency change in the public primary educational sector in Tasmania over the period could be detected. The schools‘ technical inefficiency scores iii were positively associated with students‘ suspension rates. Mothers‘ occupational status had a significant negative effect on technical inefficiency. The average technical efficiency obtained under the DEA (assuming variable returns to scale) was constant at 95% over the study period (implying no technical efficiency change). On the basis of Tobit regression results, positive effects of parental occupation on technical efficiency were also found. The number of Aboriginal and Torres Strait Islander students; students who had English as a second language; the number of disability students; students‘ absenteeism rate and that a school was classified as rural all had a negative effect on technical efficiency. Urban schools were found to be more scale efficient than rural schools. Lower scale efficiency for rural schools was due to non-optimal school size (due to remote location and low population density). The efficiency rankings of schools based on the SPF and DEA methods vary due to (i) the different ways the SPF and DEA discriminate between schools in the construction of the production frontiers, and (ii) the different methodologies SPF and DEA apply to control for environmental factors.

碩士
國立臺灣海洋大學
應用經濟研究所
97
According to Taiwan’s Fisheries Yearbook, in term of production and export, Tilapia ranks first among all species in Taiwan’s aquaculture industry. The production was 76,097 metric tons in 2007 with export exceeding all other species in Taiwan’s aquaculture industry. It is the only fishery product chosen by the Taiwan’s Council of Agriculture as one of the four flagship agricultural exports. Clearly, Tilapia is very important for Taiwan. This thesis makes use of the concept of production frontier suggested by Farrell (1957) in estimating the stochastic production frontier by FRONTIER VERSION 4.1 to analyze technical efficiency of Tilapia aquaculture and the impact of factors affecting technical efficiency on output. We determine output boundary function and the model of inefficiency as the basis for analyzing all tilapia farms. Further, this research compares technological inefficiency over various years, and estimates the value of technology inefficiency of individual farms and their spatial distribution as the basis for making suggestions to increase the value of tilapia output value. We use an integrated data over seven years in our research. In comparison with previous studies, which are based on data of a single year, we are able to study Taiwan’s tilapia industry from a broader perspective. Our result shows that translog function gives the best fits. The value for returns to scale is 1.01, indicating constant returns to scale. Technical efficiency value using translog function is 0.847 and the production error is approximately 91.7%, coming from the personal error factor. Mono-culture/poly-culture, operation scale, farming experience and the year difference all affect the technical efficiency significantly. In addition, we estimate the maximum expected output after the breeding farms overcomes the factors of technical inefficiency, allowing a more effective utilization of resources and reducing unnecessary expense. Our result may provide fishery authorities or related fishery association useful references for promoting production efficiency and achieving future production target in fish farms. Therefore, fish farms or types of aquaculture that are more efficient may be examples for the less efficient fish farms to imitate. This will be the direction for Taiwan’s aquaculture industry to improve in the future and in particular for the tilapia industry to strengthen its overall competitiveness and to promote Taiwan’s competitive advantage in the international market.

博士
國立中興大學
應用經濟學系
94
The functions of farmland include production, living, and ecology. However, different kinds of farmland encompass different functions and characteristics. The value of farmland coming from the characteristics of it may decide how to use farmland more efficiently in the future. This study classifies the farmland into suburban type of farmland, plain type of farmland and hill type of farmland. The purpose of this study is to measure the value of farmland in production, living, and ecology functions. We find that the value of production function on plain type of farmland is the highest one which is up to 268,356 NT dollars per hectare per year. The value of living function on plain type of farmland is also the highest one among these three types of farmland and it is up to 62,462 NT dollars per hectare per year. The value of ecology function on hill type of farmland is the highest one among them and it is up to 63,152 NT dollars per hectare per year. If we combine the value of all three functions on each kind of farmland, the economic value of plain type of farmland is 368,599 NT dollars per hectare per year and that of suburban type of farmland and hill type of farmland is 333,525 and 368,583 NT dollars, respectively. Consequently, we are able to obtain the combined value of each kind of farmland and to identify how to use them more efficiently. This study concludes that the plain type of farmland possesses the best production conditions and it is able to preserve rural landscape and the traditional culture the best. The hill type of farmland owns the most plentiful biological resources. Yet, the suburban type of farmland holds weak production function; thus, it can be developed to become the distribution center of agricultural products or to be released in the future. There are three kinds of functions for a farmland, but farmers usually receive the benefit of production function only. In order to raise the income of the farmer and to retrieve social fairness, we need to compensate the benefits of living and ecology functions to the farmer. If each kind of farmland can be used and planned more efficiently according to their characteristics and functions, the farmers will be able to attain more reasonable returns through the management of their farmland. The government, the farmer, and the civilians should all realize the points of view discussed in this study in order to create a sustainable development on the utilization of farmland.

Cost and production functions of US nursing homes were estimated using stochastic frontier approaches based on a 7-year panel data. Efficiency predictions from estimated frontier were then used to identify potential determinants of. This research also evaluated the effects of 1997 Balanced Budget Act on nursing home efficiencies. HCRIS-SNF was the primary database for this study and the final analysis sample size was 45,430 cost reports from 6,490 freestanding nursing homes. Total staff hours per resident-day was included in the frontier models to account for quality difference among nursing homes. Log-linear and translog functional forms were both employed in this study. A TOBIT regression model was estimated in the 2nd stage regression in addition to OLS regressions. Results from this study suggested translog functional form fits the empirical data better than the log-linear functional forms and time varying decay model is better than time invariant model. This study found that over the years US nursing homes operated at 75.4% cost inefficiency and 11.7% technical inefficiency on the average. Significant determinants of efficiency included ownership, chain affiliation, MSA, Certificate of Need/Moratoria regulation, percentage of Medicare patient days, geographic region and fiscal years. It appeared that economies of scale doesn't exist in US nursing home industry. The 1997 BBA PPS didn't seem to affect either technical or cost efficiency as of 2002. The author did observe a significant structural shift of the functions in post-BBA era
acase@tulane.edu