{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/82981"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/82981","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Investigations in Land Use Modelling: Comparing Analytical Techniques, Data Volumes, and Policy Scenarios in Sumatra","abstract":"Spatially explicit models of land use decisions have been applied to a variety of applications such as determinants of deforestation, effects of urbanization, and the influence of protected areas on land use. The models have generally been based on linear-in-parameters functional forms whose parameters are estimated from relatively small datasets. This dissertation considers the implications of linear-in-parameters models and assesses the benefits and costs of estimating the parameters for non-linear neural network specifications on large datasets as compared to the traditional linear-in-parameters specifications. The estimated models are used to simulate a variety of policy scenarios for protected areas in southern Sumatra. The parameters estimated from a large dataset are found to be substantially different than those from a small dataset. The linear-in-parameters multinomial logit and neural network specifications as estimated on the large dataset are not consistently ranked by the standard performance measures. The protected areas are found to be closely associated with the presence of evergreen forest. Removal of the protection results in deforestation for only one of the protected areas (Way Kambas National Park). Increases in population density and decreases in the cost of access index have virtually no effect on land use within protected areas. Systematic climate change results in dramatic deforestation in Way Kambas National Park as well as some minor deforestation in some of the other protected areas. The benefits of using large datasets and flexible specifications are dubious and currently outweighed by the costs for small projects. Future research should focus on obtaining better and more detailed economic data, and developing more suitable performance measures and estimation techniques.","abstract_html":"Spatially explicit models of land use decisions have been applied to a variety of applications such as determinants of deforestation, effects of urbanization, and the influence of protected areas on land use. The models have generally been based on linear-in-parameters functional forms whose parameters are estimated from relatively small datasets. This dissertation considers the implications of linear-in-parameters models and assesses the benefits and costs of estimating the parameters for non-linear neural network specifications on large datasets as compared to the traditional linear-in-parameters specifications. The estimated models are used to simulate a variety of policy scenarios for protected areas in southern Sumatra. The parameters estimated from a large dataset are found to be substantially different than those from a small dataset. The linear-in-parameters multinomial logit and neural network specifications as estimated on the large dataset are not consistently ranked by the standard performance measures. The protected areas are found to be closely associated with the presence of evergreen forest. Removal of the protection results in deforestation for only one of the protected areas (Way Kambas National Park). Increases in population density and decreases in the cost of access index have virtually no effect on land use within protected areas. Systematic climate change results in dramatic deforestation in Way Kambas National Park as well as some minor deforestation in some of the other protected areas. The benefits of using large datasets and flexible specifications are dubious and currently outweighed by the costs for small projects. Future research should focus on obtaining better and more detailed economic data, and developing more suitable performance measures and estimation techniques.","abstract_has_math":false,"creators":["Robertson, Richard D."],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Agricultural and Consumer Economics","degree_department":null,"school":null,"contributors":["Gerald C. Nelson","McNamara, Paul E."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2015,"date_issued":"2015-09-25T20:55:40Z","date_published":"2015-09-25T20:55:40Z","updated_at":"2026-07-22T22:26:20Z","subjects":["Artificial Intelligence"],"languages":["eng"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["(MiAaPQ)AAI3199126"],"render_values":[{"text":"(MiAaPQ)AAI3199126","href":null,"code":true}]}]},"links":{"outbound_url":"http://hdl.handle.net/2142/82981","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Gerald C. Nelson","McNamara, Paul E."]},{"key":"dc:creator","label":"Author","values":["Robertson, Richard D."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2015-09-25T20:55:40Z","10000-01-01","2005"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Agricultural and Consumer Economics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial Intelligence"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/82981","(MiAaPQ)AAI3199126"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Spatially explicit models of land use decisions have been applied to a variety of applications such as determinants of deforestation, effects of urbanization, and the influence of protected areas on land use. The models have generally been based on linear-in-parameters functional forms whose parameters are estimated from relatively small datasets. This dissertation considers the implications of linear-in-parameters models and assesses the benefits and costs of estimating the parameters for non-linear neural network specifications on large datasets as compared to the traditional linear-in-parameters specifications. The estimated models are used to simulate a variety of policy scenarios for protected areas in southern Sumatra. The parameters estimated from a large dataset are found to be substantially different than those from a small dataset. The linear-in-parameters multinomial logit and neural network specifications as estimated on the large dataset are not consistently ranked by the standard performance measures. The protected areas are found to be closely associated with the presence of evergreen forest. Removal of the protection results in deforestation for only one of the protected areas (Way Kambas National Park). Increases in population density and decreases in the cost of access index have virtually no effect on land use within protected areas. Systematic climate change results in dramatic deforestation in Way Kambas National Park as well as some minor deforestation in some of the other protected areas. The benefits of using large datasets and flexible specifications are dubious and currently outweighed by the costs for small projects. Future research should focus on obtaining better and more detailed economic data, and developing more suitable performance measures and estimation techniques.","Made available in DSpace on 2015-09-25T20:55:40Z (GMT). No. of bitstreams: 2 license.txt: 4848 bytes, checksum: 96035ab3f5e1c23cc7138a224ce498bd (MD5) 3199126.pdf: 3728982 bytes, checksum: d1e22a7c35d18eade176eac2ac77a48b (MD5) Previous issue date: 2005","Embargo set by: Seth Robbins for item 84262 Lift date: Forever Reason: Restricted to the U of I community idenfinitely during batch ingest of legacy ETDs","Restricted to the U of I community idenfinitely during batch ingest of legacy ETDs","U of I Only","121 p.","Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2005."]},{"key":"dc:title","label":"Title","values":["Investigations in Land Use Modelling: Comparing Analytical Techniques, Data Volumes, and Policy Scenarios in Sumatra"]}]}],"canonical_facts":{"dc:contributor":["Gerald C. Nelson","McNamara, Paul E."],"dc:creator":["Robertson, Richard D."],"dc:date":["2015-09-25T20:55:40Z","10000-01-01","2005"],"dc:description":["Spatially explicit models of land use decisions have been applied to a variety of applications such as determinants of deforestation, effects of urbanization, and the influence of protected areas on land use. The models have generally been based on linear-in-parameters functional forms whose parameters are estimated from relatively small datasets. This dissertation considers the implications of linear-in-parameters models and assesses the benefits and costs of estimating the parameters for non-linear neural network specifications on large datasets as compared to the traditional linear-in-parameters specifications. The estimated models are used to simulate a variety of policy scenarios for protected areas in southern Sumatra. The parameters estimated from a large dataset are found to be substantially different than those from a small dataset. The linear-in-parameters multinomial logit and neural network specifications as estimated on the large dataset are not consistently ranked by the standard performance measures. The protected areas are found to be closely associated with the presence of evergreen forest. Removal of the protection results in deforestation for only one of the protected areas (Way Kambas National Park). Increases in population density and decreases in the cost of access index have virtually no effect on land use within protected areas. Systematic climate change results in dramatic deforestation in Way Kambas National Park as well as some minor deforestation in some of the other protected areas. The benefits of using large datasets and flexible specifications are dubious and currently outweighed by the costs for small projects. Future research should focus on obtaining better and more detailed economic data, and developing more suitable performance measures and estimation techniques.","Made available in DSpace on 2015-09-25T20:55:40Z (GMT). No. of bitstreams: 2 license.txt: 4848 bytes, checksum: 96035ab3f5e1c23cc7138a224ce498bd (MD5) 3199126.pdf: 3728982 bytes, checksum: d1e22a7c35d18eade176eac2ac77a48b (MD5) Previous issue date: 2005","Embargo set by: Seth Robbins for item 84262 Lift date: Forever Reason: Restricted to the U of I community idenfinitely during batch ingest of legacy ETDs","Restricted to the U of I community idenfinitely during batch ingest of legacy ETDs","U of I Only","121 p.","Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2005."],"dc:identifier":["http://hdl.handle.net/2142/82981","(MiAaPQ)AAI3199126"],"dc:language":["eng"],"dc:subject":["Artificial Intelligence"],"dc:title":["Investigations in Land Use Modelling: Comparing Analytical Techniques, Data Volumes, and Policy Scenarios in Sumatra"],"dc:type":["text"],"thesis:degree_discipline":["Agricultural and Consumer Economics"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:26:20Z"}