1 / 17

Landbouweconomie , Coupure Links 653, 9000 Gent

FACULTEITBIO-INGENIEURSWETENSCHAPPEN. Sub-vector Efficiency Analysis in Chance Constrained Stochastic DEA: An Application to Irrigation Water Use in Krishna River Basin, India Prakashan Chellattan Veettil et. al. Department of Agricultural Economics, Ghent University, Belgium.

kera
Download Presentation

Landbouweconomie , Coupure Links 653, 9000 Gent

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. FACULTEITBIO-INGENIEURSWETENSCHAPPEN Sub-vector Efficiency Analysis in Chance Constrained Stochastic DEA: An Application to Irrigation Water Use in Krishna River Basin, IndiaPrakashan Chellattan Veettil et. al.Department of Agricultural Economics, Ghent University, Belgium 122nd EAAE Seminar: Ancona (Italy) (17-18 Feb 2011) Landbouweconomie, Coupure Links 653, 9000 Gent e-mail: Prakashan.ChellattanVeettil@UGent.be

  2. Problem Statement Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Setting the scene • Why we need to analyze the Performance of productive units? • Maximize Profit or Minimize cost • Scarce resources • Better allocation • We need an appropriate methodology: • Different approaches • Parametric Approach: e.g. SFA • Non-Parametric Approach: e.g. DEA 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  3. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Parametric Approach Non-Parametric Approach • Parametric Approach • Functional form for production relation • Differentiate between inefficiency and noise. • Statistical foundation for testing (ML and LRT) • Limitations • A priori justification for • 1. Distribution of inefficiency terms • 2. Functional form for frontier 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  4. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Parametric Approach Non-Parametric Approach Non-Parametric Approach: e.g.: Data Envelopment Analysis (DEA) Flexible functional form for input-output relations Measures relative efficiency of a DMU 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  5. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Determinist DEA Stochastic DEA In-efficiency Linear Programming Formulation 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  6. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Determinist DEA Stochastic DEA Weaknesses of standard DEA approaches No assumption of in-efficiency across DMUs No Noise assumption: Measurement errorcan cause significant problems Relative efficiency: DEA does not measure "absolute" efficiency; but asymptotically equal Sensitive to outliers and extreme values 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  7. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Motivation 1 Motivation 2 • Motivation 1: Stochastic DEA • Most of the production relationships are not deterministic • Uncertainties in agricultural production system (climate, input, price, management uncertainties etc.) • Data is noisy and introducing Noise in DEA is a challenging task • Further Restriction is demanded • Robust to outliers and extreme values • Combining the advantages of SFA with Non-parametric method (DEA) 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  8. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Motivation 1 Motivation 2 Output Oriented SDEA frontier Input Oriented SDEA frontier 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  9. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Motivation 1 Motivation 2 Stochastic Efficiency The maximum probability of the existence of dominating DMU is α, then the DMU is called α-stochastically efficient Where α is the probability limit of stochasticity 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  10. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Motivation 1 Motivation 2 Stochastic DEA formulation 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  11. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Motivation 1 Motivation 2 • Motivation 2: Sub-vector Stochastic DEA • Sub-vector of inputs contracted for output production • For example: Agricultural production system • Water limiting factor in arid climate • Agricultural production relationships sub-optimal if we do not take Water use efficiency (sub-vector) 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  12. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Determinist DEA Stochastic DEA Stochastic DEA formulation: sub-vector case 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  13. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Output efficiency Sub-vector Efficiency Band of soft frontier accommodates more farms as stochastically efficient (fraction of farmers on stochastic output frontier is higher) The output efficiency in SDEA is higher than DEA under both frameworks 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  14. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Output efficiency Sub-vector Efficiency Difference in output efficiency scores between DDEA and SDEA Difference in Water Use efficiency scores between DDEA and SDEA 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  15. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Output efficiency Sub-vector Efficiency Similar pattern as that of output efficiency scores DEA model: those farmers who are efficient producers (output efficiency) are also efficient in water use (sub-vector efficiency) SDEA someefficient producers are not efficient in water use 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  16. Introduction Efficiency Analysis Data Envelopment Analysis (DEA) Result and Conclusion Conclusion Conclusion Introducing Stochasticity into Non-parametric Efficiency analysis is complex But the limitation of DEA to incorporate noise often pays higher costs The influence of fewer farms (possible outliers??) on frontier can be reduced A better representation of frontier technology POLICY: Not all efficient producers are efficient water users. Re-allocation of resources are possible through input oriented policy rather than production oriented ones. 122nd EAAE Seminar, Ancona: 17-18 Feb 2011 e-mail: Prakashan.ChellattanVeettil@UGent.be

  17. FACULTEITBIO-INGENIEURSWETENSCHAPPEN THANK YOU Landbouweconomie, Coupure Links 653, 9000 Gent e-mail: Prakashan.ChellattanVeettil@UGent.be

More Related