{"corpus_id":236365554,"paper_sha":"c020d9abf9f359acaf12f34a16a17ae74d0912ca","doi":"10.1080/00036846.2021.1904124","arxiv_id":null,"pmid":null,"pmcid":null,"mag_id":3161586235,"dblp_id":null,"acl_id":null,"title":"SFA robustness to violated distributional assumptions: theory, simulations and empirical evidence","year":2021,"publication_date":"2021-05-18","venue":"","journal":{"name":"Applied Economics","pages":"4544 - 4559","volume":"53"},"journal_issn":null,"journal_title":null,"publication_types":[],"pubmed_pub_types":null,"s2_fields_of_study":["Economics"],"reference_count":64,"citation_count":3,"influential_citation_count":0,"is_open_access":false,"arxiv_categories":null,"arxiv_license":null,"arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":null,"s2_open_access_landing_url":null,"s2_open_access_license":null,"s2_open_access_status":null,"pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"ABSTRACT The well-known biases caused by taking the logarithm of a heteroskedastic dependent variable are a potential threat in stochastic frontier analyses (SFA), which are often estimated in log-linear form. This paper shows that while these biases can indeed be substantial, they do not affect the identification of best and worst performing firms, which is often more important than parameters of the frontier function. The efficiency ranks remain correctly identified except for special cases which are easily detectable. Furthermore, we document that the popular normal-truncated-normal SFA model with time-invariant efficiency shows a notable robustness to the mis-specification of the distribution of efficiency terms. The reason for this property is the model’s similarity to Gaussian pseudo-maximum likelihood estimator. Simulations are used to illustrate these properties and to show that the recovery of the correct productivity ranks is feasible even with very short panels. Empirical exercises show that using short panels is practicable in real datasets.","claims":[{"public_id":"cl_990f57ad5758ab2af576b77390406be2","status":"active","text":"Biases from taking the logarithm of a heteroskedastic dependent variable in stochastic frontier analyses do not affect the identification of best and worst performing firms, which is often more important than parameters of the frontier function.","confidence":0.9,"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_990f57ad5758ab2af576b77390406be2"},{"public_id":"cl_a99926209980ec1a672c02818ceac275","status":"active","text":"Efficiency ranks remain correctly identified except for special cases which are easily detectable.","confidence":0.85,"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_a99926209980ec1a672c02818ceac275"},{"public_id":"cl_f1dc77fd54be97eef3550664194c03b6","status":"active","text":"Empirical exercises show that using short panels is practicable in real datasets.","confidence":0.8,"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_f1dc77fd54be97eef3550664194c03b6"},{"public_id":"cl_633d19cbc4f4bcbf3d5a31467e3edbd1","status":"active","text":"Recovery of correct productivity ranks is feasible even with very short panels, as shown by simulations.","confidence":0.85,"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_633d19cbc4f4bcbf3d5a31467e3edbd1"},{"public_id":"cl_734abebc4bc2a522ebd8bcfb33fb08a8","status":"active","text":"The popular normal-truncated-normal SFA model with time-invariant efficiency shows notable robustness to mis-specification of the distribution of efficiency terms due to its similarity to Gaussian pseudo-maximum likelihood estimator.","confidence":0.9,"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_734abebc4bc2a522ebd8bcfb33fb08a8"}],"concepts":[{"public_id":"co_3a60aae192b819898acc6368aea14f19","status":"active","name":"logarithm of heteroskedastic dependent variable","description":"The transformation of a dependent variable that exhibits heteroskedasticity by taking its logarithm, which can introduce biases in stochastic frontier analyses.","types":["transformation","source of bias"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_3a60aae192b819898acc6368aea14f19"},{"public_id":"co_4069ba7f64d2884ae5d0f7970dbc2afd","status":"active","name":"time-invariant efficiency","description":"An assumption that firm-level efficiency does not change over time in the panel data setting.","types":["assumption"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_4069ba7f64d2884ae5d0f7970dbc2afd"},{"public_id":"co_49e46cef9c8cdc0e2a13001fc0559f4d","status":"active","name":"Gaussian pseudo-maximum likelihood estimator","description":"An estimator that maximizes a likelihood based on a Gaussian (normal) distribution even when the true distribution may differ, providing robustness to distributional mis-specification.","types":["estimator"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_49e46cef9c8cdc0e2a13001fc0559f4d"},{"public_id":"co_526788a016c577d29c80fd857743e6b8","status":"active","name":"identification of best and worst performing firms","description":"The ability to correctly rank firms at the top and bottom of the performance distribution, which is argued to be robust to biases from log-linear estimation.","types":["objective"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_526788a016c577d29c80fd857743e6b8"},{"public_id":"co_5846a7941d192b07c2ccaa9a3bb82d1a","status":"active","name":"short panels","description":"Panel datasets with few time periods per unit, which are shown to be feasible for recovering productivity ranks in both simulations and empirical exercises.","types":["data characteristic"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_5846a7941d192b07c2ccaa9a3bb82d1a"},{"public_id":"co_5b02fe294c23a7b7141e682f692a0fe3","status":"active","name":"normal-truncated-normal SFA model","description":"A stochastic frontier model that assumes a normal distribution for the random error and a truncated normal distribution for the inefficiency term.","types":["model"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_5b02fe294c23a7b7141e682f692a0fe3"},{"public_id":"co_8ae9f5c84dcd5abd4ae9bc4465dce45e","status":"active","name":"efficiency ranks","description":"The ordering of firms by their estimated technical efficiency, which remains correctly identified except in special detectable cases.","types":["output"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_8ae9f5c84dcd5abd4ae9bc4465dce45e"},{"public_id":"co_8cc33396bf3c8f5cdd74f7d73c67c9f7","status":"active","name":"stochastic frontier analyses","description":"Econometric methods for estimating production or cost frontiers that incorporate a stochastic error term, often estimated in log-linear form.","types":["method"],"aliases":["SFA"],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_8cc33396bf3c8f5cdd74f7d73c67c9f7"},{"public_id":"co_9eddb1654e9c19a3e4126b6686b687ce","status":"active","name":"productivity ranks","description":"The ordering of firms by their productivity levels, which can be recovered even with very short panels according to simulations.","types":["output"],"aliases":[],"contributors":[{"id":32,"public_id":"7c402c1b98","public_label":"뀨 (7c402c1b98)","roles":["extraction"],"url":"https://sah.borca.ai/u/7c402c1b98"},{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["review"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_9eddb1654e9c19a3e4126b6686b687ce"}],"external_ids":{"DOI":"10.1080/00036846.2021.1904124","ArXiv":null,"PubMed":null,"PubMedCentral":null,"MAG":3161586235,"DBLP":null,"ACL":null},"open_access":{"is_open_access":false,"pdf_url":null,"landing_url":"https://sah.borca.ai/papers/236365554","source":null,"pdf_url_source":null,"license":null,"reason":"pdf_url_not_indexed"},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":false,"full_text_source":null,"count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":634322,"paper_uid":"f4381339-0154-493d-a20a-fbb2252b4ee1","canonical_identity":{"paper_id":634322,"paper_uid":"f4381339-0154-493d-a20a-fbb2252b4ee1","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/236365554"}