ECONOMETRIC METHODS FOR AGRICULTURAL AND APPLIED ECONOMICS

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 0222137
• Project Start Date: Apr 22, 2010
• Project End Date: Apr 21, 2015
• Program Code: [(N/A)]- (N/A)

Recipient Organization
TEXAS A&M UNIVERSITY
750 AGRONOMY RD STE 2701
COLLEGE STATION,TX 77843-0001

Performing Department
Agri Economics

Non Technical Summary
Econometric analysis of agricultural and applied economics problem is an important part of economic analysis. Guided by empirical observations and economic theories, econometric analysis can provide crucial insight into real world problems. It serves several functions, including validation of economic theories, quantification of theoretical predication and provision of parameters for predictions and simulations. With the advance of modern computing technology, economists are increasingly less constrained in their econometric investigation by the complexity of model and estimations, and/or lack of computing power. The past decades have witnessed considerable advance in new econometric methods that are computationally intensive. For instance, nonparametric estimation, bootstrap inference, multivariate analysis and functional data analysis, have been adapted by applied economists in their investigation of a wide range of economic problems.

Animal Health Component

(N/A)

Research Effort Categories

Basic

(N/A)

Applied

(N/A)

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
601	6110	3010	60%
609	6110	2090	40%

Knowledge Area
601 - Economics of Agricultural Production and Farm Management; 609 - Economic Theory and Methods;

Subject Of Investigation
6110 - Economy of the United States and sectors thereof;

Field Of Science
2090 - Statistics, econometrics, and biometrics; 3010 - Economics;

Keywords

agricultural economics

econometrics

information-theoretic methods

nonparametric econometrics

Goals / Objectives
This proposal has two main objectives: 1. Construction and investigation of new econometric methods motivated by agricultural and applied economics issues; 2. Application to real world problems.The successful implementation of the proposed methods outlined above can find wide-ranged applications in agricultural and applied economics. For example, the PI plans to pursue applications in the following areas rigorously. (1) Large scale demand analysis. Modern day demand analysis is often featured by large scale data sets such as scanner data. However, for individual items, data scarcity and truncation pose severe difficulties for consistent demand estimation. The proposed multivariate estimation method with flexible distributional assumption is hoped to provide a viable solution to this important problem. (2) Multivariate risk analysis. The recent financial market recession signifies the importance of understanding the joint movement of financial assets, specially the co-movements of extreme financial events. A multivariate density estimation and optimal subset selection mechanism would provide an efficient estimation for the joint distribution of financial asset returns and a vehicle to understand the co-movements of their tails. In particular, the copula method will be used to analyze the dependence among asset returns. (3) Modeling of crop yields. It has been suggested that the climate change of the past few decades has a non-negligible impact on crop yields. Thus it is very important to understand how climate conditions affect crop yields. While daily climate data are readily available, crop yield information is typically reported at on an annual basis. One possible solution is to process the climate data as functional data for each crop/year, and then use principle component analysis method to investigate their impacts. In particular, the nonparametric smoothing spline method will be suitable for this task due to its adaptiveness and versatility. (4) Multivariate welfare analysis. Conventional welfare inference based on one single attribute (such as income or consumption) is inadequate as it only covers one aspect of human welfare. However, multivariate welfare comparison is not directly feasible without fully specifying agents' individual utility functions or equivalently a multivariate social welfare function. Alternatively, one can use the method of stochastic dominance to undertake welfare comparison. This approach relaxes the requirement on welfare functions, but it is difficult in terms of implementation. The PI plans to use nonparametric methods to derive tests for higher dimensional stochastic dominance. A solution to this problem will provide an important tool to both economists and policy makers.

Project Methods
One main objective of this proposal is to combine information-theoretic methods with nonparametric analysis to arrive at flexible yet practically useful estimators. This is indeed an important research area the PI has been working on for the past few years. For example, in Wu and Perloff (2007), the conventional maximum entropy method is generalized to deal with interval data. This method is used to estimate the income distribution of China, whose income information is only available in the form of summary statistics by intervals. Wu (forthcoming) further extend the maximum entropy density method in two dimensions: (1) the number of moments is allowed to increase with sample size at a proper rate to achieve consistency; (2) it is extended to high dimensional space. This method is shown to produce superior estimator for copula density estimation. The PI plans to further his investigation of information-theoretic and nonparametric methods in the following directions: (1) Design of optimal subset selection methods for multivariate analysis. This is of particular importance, without which the estimation will be hampered by severe curse of dimensionality. (2) Feasible implementation of generalized GMM methods, including the empirical likelihood, exponential tilting and continuously updating estimators. These estimators are known for their higher order efficiency. However, their adaptation by applied researchers is rare due to the numerical difficulty in terms of implementation. A generalization of these estimators in the line of generalized maximum entropy estimation by incorporating explicitly noise or penalty terms may be a possible solution to this problem. (3) Multivariate censoring or truncating models with flexible distributional assumptions. The classical Heckman two-step procedure cannot be generalized easily into higher-than-two dimensional case. Thus a flexible model that can account for agents' discrete choice and the subsequent continuous outcome would be a very useful addition to economists' toolbox.

Progress 04/22/10 to 04/21/15

Outputs
Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Scientific publications.

Publications

Progress 10/01/13 to 09/30/14

Outputs
Target Audience: This proposal has two main objectives: 1. Construction and investigation of new econometric methods motivated by agricultural and applied economics issues; 2. Application to real world problems.The successful implementation of the proposed methods outlined above can find wide-ranged applications in agricultural and applied economics. For example, the PI plans to pursue applications in the following areas rigorously. (1) Large scale demand analysis. Modern day demand analysis is often featured by large scale data sets such as scanner data. However, for individual items, data scarcity and truncation pose severe difficulties for consistent demand estimation. The proposed multivariate estimation method with flexible distributional assumption is hoped to provide a viable solution to this important problem. (2) Multivariate risk analysis. The recent financial market recession signifies the importance of understanding the joint movement of financial assets, specially the co-movements of extreme financial events. A multivariate density estimation and optimal subset selection mechanism would provide an efficient estimation for the joint distribution of financial asset returns and a vehicle to understand the co-movements of their tails. In particular, the copula method will be used to analyze the dependence among asset returns. (3) Modeling of crop yields. It has been suggested that the climate change of the past few decades has a non-negligible impact on crop yields. Thus it is very important to understand how climate conditions affect crop yields. While daily climate data are readily available, crop yield information is typically reported at on an annual basis. One possible solution is to process the climate data as functional data for each crop/year, and then use principle component analysis method to investigate their impacts. In particular, the nonparametric smoothing spline method will be suitable for this task due to its adaptiveness and versatility. (4) Multivariate welfare analysis. Conventional welfare inference based on one single attribute (such as income or consumption) is inadequate as it only covers one aspect of human welfare. However, multivariate welfare comparison is not directly feasible without fully specifying agents' individual utility functions or equivalently a multivariate social welfare function. Alternatively, one can use the method of stochastic dominance to undertake welfare comparison. This approach relaxes the requirement on welfare functions, but it is difficult in terms of implementation. The PI plans to use nonparametric methods to derive tests for higher dimensional stochastic dominance. A solution to this problem will provide an important tool to both economists and policy makers. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Training of graduate students. How have the results been disseminated to communities of interest? Profession publications and conferences; seminars What do you plan to do during the next reporting period to accomplish the goals? continue current work on related projects

Impacts
What was accomplished under these goals? The methods investigated in the above mentioned papers were used to estimate agricultural production functions, impact of climate changes on crop yields distributions and public health, and U.S. waterway transportation systems.

Publications

• Type: Book Chapters Status: Accepted Year Published: 2014 Citation: D. Bessler, B. McCarl, X. Wu and H.A. Love. Quantitative Methods in Agricultural Economics. In Encyclopedia of Agriculture Science, Academic Press, forthcoming.
• Type: Journal Articles Status: Published Year Published: 2014 Citation: J.E. Mu, B.A. McCarl, X. Wu, and M. Ward. Climate change and the risk of highly pathogenic avian influenza outbreaks in birds. British Journal of Environment and Climate Change, 4:166185, 2014.
• Type: Journal Articles Status: Published Year Published: 2014 Citation: T. Dang, D. Leatham, B. McCarl, and X. Wu. Measuring efficiency within the farm credit system. Agricultural Finance Review, 74:3854, 2014.
• Type: Journal Articles Status: Accepted Year Published: 2015 Citation: Y. Gao, Y. Zhang, and X. Wu. Penalized exponential series estimation of copula densities with an application to intergenerational dependence of body mass index. Empirical Economics, accepted.
• Type: Journal Articles Status: Accepted Year Published: 2015 Citation: Q. Li, X. Wu, and D. Zhang. A data-driven smooth test of symmetry. Journal of Econometrics, accepted.
• Type: Journal Articles Status: Accepted Year Published: 2015 Citation: J. Lin and X. Wu. Smooth tests of copula specifications. Journal of Business and Economic Statistics,accepted.
• Type: Journal Articles Status: Accepted Year Published: 2015 Citation: K. Wen and X. Wu. An improved transformation-based kernel estimator of densities on the unit interval. Journal of the American Statistical Association, accepted.
• Type: Journal Articles Status: Accepted Year Published: 2015 Citation: M. Chang and X. Wu. Transformation-based nonparametric estimation of multivariate densities. Journal of Multivariate Analysis, accepted.

Progress 01/01/13 to 09/30/13

Outputs
Target Audience: This proposal has two main objectives: 1. Construction and investigation of new econometric methods motivated by agricultural and applied economics issues; 2. Application to real world problems.The successful implementation of the proposed methods outlined above can find wide-ranged applications in agricultural and applied economics. For example, the PI plans to pursue applications in the following areas rigorously. (1) Large scale demand analysis. Modern day demand analysis is often featured by large scale data sets such as scanner data. However, for individual items, data scarcity and truncation pose severe difficulties for consistent demand estimation. The proposed multivariate estimation method with flexible distributional assumption is hoped to provide a viable solution to this important problem. (2) Multivariate risk analysis. The recent financial market recession signifies the importance of understanding the joint movement of financial assets, specially the co-movements of extreme financial events. A multivariate density estimation and optimal subset selection mechanism would provide an efficient estimation for the joint distribution of financial asset returns and a vehicle to understand the co-movements of their tails. In particular, the copula method will be used to analyze the dependence among asset returns. (3) Modeling of crop yields. It has been suggested that the climate change of the past few decades has a non-negligible impact on crop yields. Thus it is very important to understand how climate conditions affect crop yields. While daily climate data are readily available, crop yield information is typically reported at on an annual basis. One possible solution is to process the climate data as functional data for each crop/year, and then use principle component analysis method to investigate their impacts. In particular, the nonparametric smoothing spline method will be suitable for this task due to its adaptiveness and versatility. (4) Multivariate welfare analysis. Conventional welfare inference based on one single attribute (such as income or consumption) is inadequate as it only covers one aspect of human welfare. However, multivariate welfare comparison is not directly feasible without fully specifying agents' individual utility functions or equivalently a multivariate social welfare function. Alternatively, one can use the method of stochastic dominance to undertake welfare comparison. This approach relaxes the requirement on welfare functions, but it is difficult in terms of implementation. The PI plans to use nonparametric methods to derive tests for higher dimensional stochastic dominance. A solution to this problem will provide an important tool to both economists and policy makers. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? OUTPUTS: Several working papers on estimation of production functions, crop yield distributions and copula density function estimations. Revised papers: "Climate Change Influences on Agricultural Research Productivity," (with X. Villavicencio, B. McCarl and W. Huffman), revised and resubmitted, Climatic Change "Climate Change and the Risk of Highly Pathogenic Avian Influenza H5N1 Outbreaks," (with J. Mu, B. McCarl and M. Ward), revision requested from Climatic Research How have the results been disseminated to communities of interest? professional publications and conferences; seminars What do you plan to do during the next reporting period to accomplish the goals? continue current work on related projects

Impacts
What was accomplished under these goals? The methods investigated in the above mentioned papers were used to estimate agricultural production functions, impact of climate changes on crop yields distributions and public health, and U.S. waterway transportation systems.

Publications

• Type: Journal Articles Status: Published Year Published: 2013 Citation: "Global Joint Distribution of Income and Health," (with A. Savvides and T. Stengos), In J. Ma and M. Wohar, editors, Recent Advances in Estimating Nonlinear Models with Applications in Economics and Finance, Springer Books, 2013
• Type: Journal Articles Status: Published Year Published: 2013 Citation: "Climate Change Influences on Agricultural Research Productivity," (with X. Villavicencio, B. McCarl, and W. Huffman), Climatic Change, 2013
• Type: Journal Articles Status: Published Year Published: 2013 Citation: Dang, T., D.J. Leatham, B.A. McCarl, and X.M. Wu, Measuring efficiency within the farm credit system. Agricultural Finance Review, forthcoming, 2013

Progress 01/01/12 to 12/31/12

Outputs
OUTPUTS: Several working papers on estimation of production functions, crop yield distributions and copula density function estimations. Revised papers: "Climate Change Influences on Agricultural Research Productivity," (with X. Villavicencio, B. McCarl and W. Huffman), revised and resubmitted, Climatic Change "Climate Change and the Risk of Highly Pathogenic Avian Influenza H5N1 Outbreaks," (with J. Mu, B. McCarl and M. Ward), revision requested from Climatic Research PARTICIPANTS: Nothing significant to report during this reporting period. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
The methods investigated in the above mentioned papers were used to estimate agricultural production functions, impact of climate changes on crop yields distributions and public health, and U.S. waterway transportation systems.

Publications

• X. Wu, A. Savvides, and T. Stengos. Global joint distribution of income and health. In J. Ma and M. Wohar, editors, Recent Advances in Estimating Nonlinear Models With Applications In Economics and Finance. Springer Books, 2012.
• Y. Liu, X. Wu, and W. Zeng. Detecting statistical abnormality by combining benfords law and panel data models. Statistical Research (in Chinese), 2012.

Progress 01/01/11 to 12/31/11

Outputs
OUTPUTS: During the year of 2011, the PI conducted research in various areas of applied economics and econometrics. The main outputs of this year include development of new econometrics methods is estimation of production function and/or cost function under shape restrictions implied by economic theories. The PI has also conducted two projects related to the statistical analysis of US waterway transportation system and their impact on US rural community. PARTICIPANTS: Not relevant to this project. TARGET AUDIENCES: The target audiences of my research include the agricultural economists, applied economists and econometricians. PROJECT MODIFICATIONS: Not relevant to this project.

Impacts
Several papers and two reports to the University Transportation Center of Mobility were completed during 2011. The method developed by the PI and the public available computer code has generated interest from users that include not only economists, but researchers in other fields of science. The method proposed by the PI has been used by other researchers and led to publications in professional journals. The PI presented the methodology and related findings in several professional conferences.

Publications

• Egbendewe-Mondzozo A., Musumba M., McCarl B.A. and X. Wu. Climate Change and Vector-borne Diseases: An Economic Impact Analysis of Malaria in Africa. International Journal of Environmental Research and Public Health, 8(3), 913-930, 2011.
• G. Gustavsen, R. Nayga and X. Wu. Obesity and Moral Hazard in Demands for Visits to Physicians. Contemporary Economic Policy, 29, 620-633, 2011.
• X. Wu and S. Wang. Information theoretic asymptotic approximations for distributions of statistics. Scandinavian Journal of Statistics, 38,130-146, 2011.

Progress 01/01/10 to 12/31/10

Outputs
OUTPUTS: During the year of 2010, the PI conducted research in various areas of applied economics and econometrics. The main outputs of this year include development of new econometrics methods than can be used to analyze multivariate data, especially for copula densities, which are used extensively in the financial sector. Another area of the PI's active research is the modeling and estimation of production function and/or cost function under shape restrictions implied by economic theories. Several computer programs developed during the research are available at the PI's website. PARTICIPANTS: Meng-Shiuh Chang, Ke Li. Both phd students in the department of Agricultural Economics. TARGET AUDIENCES: The target audiences of my research include the agricultural economists, applied economists and econometricians. PROJECT MODIFICATIONS: No major changes.

Impacts
The method developed by the PI and the public available computer code has generated interest from users that include not only economists, but researchers in other fields of science. The method proposed by the PI has been used by other researchers and led to publications in professional journals. The PI presented the methodology and related findings in several professional conferences.

Publications

• [1] X. Wu and S. Wang. Information theoretic asymptotic approximations for distributions of statistics. Scandinavian Journal of Statistics, April, 2010.
• [2] X. Wu. Exponential series estimator of multivariate densities. Journal of Econometrics, 156:354-366, 2010.