{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,22]],"date-time":"2026-01-22T15:20:37Z","timestamp":1769095237567,"version":"3.49.0"},"reference-count":45,"publisher":"Springer Science and Business Media LLC","issue":"3","license":[{"start":{"date-parts":[[2021,2,22]],"date-time":"2021-02-22T00:00:00Z","timestamp":1613952000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,2,22]],"date-time":"2021-02-22T00:00:00Z","timestamp":1613952000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Vienna University of Economics and Business"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Comput Stat"],"published-print":{"date-parts":[[2021,9]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Relative risks are estimated to assess associations and effects due to their ease of interpretability, e.g., in epidemiological studies. Fitting log-binomial regression models allows to use the estimated regression coefficients to directly infer the relative risks. The estimation of these models, however, is complicated because of the constraints which have to be imposed on the parameter space. In this paper we systematically compare different optimization algorithms to obtain the maximum likelihood estimates for the regression coefficients in log-binomial regression. We first establish under which conditions the maximum likelihood estimates are guaranteed to be finite and unique, which allows to identify and exclude problematic cases. In simulation studies using artificial data we compare the performance of different optimizers including solvers based on the augmented Lagrangian method, interior-point methods including a conic optimizer, majorize-minimize algorithms, iteratively reweighted least squares and expectation-maximization algorithm variants. We demonstrate that conic optimizers emerge as the preferred choice due to their reliability, lack of requirement to tune hyperparameters and speed.<\/jats:p>","DOI":"10.1007\/s00180-021-01084-5","type":"journal-article","created":{"date-parts":[[2021,2,22]],"date-time":"2021-02-22T09:04:54Z","timestamp":1613984694000},"page":"1721-1754","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["A comparison of optimization solvers for log binomial regression including conic programming"],"prefix":"10.1007","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3983-9773","authenticated-orcid":false,"given":"Florian","family":"Schwendinger","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7265-4773","authenticated-orcid":false,"given":"Bettina","family":"Gr\u00fcn","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4198-9911","authenticated-orcid":false,"given":"Kurt","family":"Hornik","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2021,2,22]]},"reference":[{"issue":"1","key":"1084_CR1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1093\/biomet\/71.1.1","volume":"71","author":"A Albert","year":"1984","unstructured":"Albert A, Anderson JA (1984) On the existence of maximum likelihood estimates in logistic regression models. Biometrika 71(1):1\u201310","journal-title":"Biometrika"},{"issue":"1","key":"1084_CR2","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1002\/bimj.200410165","volume":"48","author":"L Blizzard","year":"2006","unstructured":"Blizzard L, Hosmer W (2006) Parameter estimation and goodness-of-fit in log binomial regression. Biom J 48(1):5\u201322. https:\/\/doi.org\/10.1002\/bimj.200410165","journal-title":"Biom J"},{"key":"1084_CR3","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511804441","volume-title":"Convex optimization","author":"S Boyd","year":"2004","unstructured":"Boyd S, Vandenberghe L (2004) Convex optimization. Cambridge University Press, New York"},{"key":"1084_CR4","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9781107279667","volume-title":"Optimization models","author":"GC Calafiore","year":"2014","unstructured":"Calafiore GC, El Ghaoui L (2014) Optimization models. Cambridge University Press, Cambridge. https:\/\/doi.org\/10.1017\/CBO9781107279667"},{"key":"1084_CR5","unstructured":"Chares PR (2009) Cones and interior-point algorithms for structured convex optimization involving powers and exponentials. PhD thesis, Universit\u00e9 Catholique de Louvain. http:\/\/dial.uclouvain.be\/pr\/boreal\/en\/object\/boreal%3A28538\/datastream\/PDF_01\/view"},{"issue":"7136","key":"1084_CR6","doi-asserted-by":"publisher","first-page":"989","DOI":"10.1136\/bmj.316.7136.989","volume":"316","author":"HT Davies","year":"1998","unstructured":"Davies HT, Crombie IK, Tavakoli M (1998) When can odds ratios mislead? BMJ (Clin Res Ed) 316(7136):989\u2013991. https:\/\/doi.org\/10.1136\/bmj.316.7136.989","journal-title":"BMJ (Clin Res Ed)"},{"key":"1084_CR7","first-page":"1","volume":"29","author":"BB De Andrade","year":"2011","unstructured":"De Andrade BB, Carabin H (2011) On the estimation of relative risks via log binomial regression. Revista Brasileira de Biometria 29:1\u201315","journal-title":"Revista Brasileira de Biometria"},{"issue":"23","key":"1084_CR8","doi-asserted-by":"publisher","first-page":"5750","DOI":"10.1080\/03610926.2017.1402045","volume":"47","author":"BB de Andrade","year":"2018","unstructured":"de Andrade BB, de Leon Andrade JM (2018) Some results for maximum likelihood estimation of adjusted relative risks. Commun Stat Theory Methods 47(23):5750\u20135769. https:\/\/doi.org\/10.1080\/03610926.2017.1402045","journal-title":"Commun Stat Theory Methods"},{"key":"1084_CR9","unstructured":"De\u00a0Leeuw J, Heiser W (1977) Convergence of correction matrix algorithms for multidimensional scaling. In: Lingoes J (ed) Geometric representations of relational data. Mathesis Press, Ann Arbor, chap\u00a032, pp 735\u2013753. http:\/\/deleeuwpdx.net\/janspubs\/1977\/chapters\/deleeuw_heiser_C_77.pdf"},{"key":"1084_CR10","doi-asserted-by":"publisher","unstructured":"Domahidi A (2013) Methods and tools for embedded optimization and control. PhD thesis, ETH Zurich. https:\/\/doi.org\/10.3929\/ethz-a-010010483","DOI":"10.3929\/ethz-a-010010483"},{"key":"1084_CR11","doi-asserted-by":"crossref","unstructured":"Domahidi A, Chu E, Boyd S (2013) ECOS: an SOCP solver for embedded systems. In: European control conference (ECC), pp 3071\u20133076. https:\/\/web.stanford.edu\/~boyd\/papers\/pdf\/ecos_ecc.pdf","DOI":"10.23919\/ECC.2013.6669541"},{"issue":"9","key":"1084_CR12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18637\/jss.v086.i09","volume":"86","author":"M Donoghoe","year":"2018","unstructured":"Donoghoe M, Marschner I (2018) logbin: an R package for relative risk regression using the log-binomial model. J Stat Softw 86(9):1\u201322 10.18637\/jss.v086.i09","journal-title":"J Stat Softw"},{"key":"1084_CR13","volume-title":"Nonlinear programming: sequential unconstrained minimization techniques","author":"AV Fiacco","year":"1968","unstructured":"Fiacco AV, McCormick GP (1968) Nonlinear programming: sequential unconstrained minimization techniques. Wiley, New York"},{"key":"1084_CR14","doi-asserted-by":"crossref","unstructured":"Holcomb WL, Chaiworapongsa T, Luke DA, Burgdorf KD (2001) An odd measure of risk: use and misuse of the odds ratio. Obstet Gynecol 98(4). https:\/\/journals.lww.com\/greenjournal\/Fulltext\/2001\/10000\/An_Odd_Measure_of_Risk__Use_and_Misuse_of_the_Odds.28.aspx","DOI":"10.1097\/00006250-200110000-00028"},{"key":"1084_CR15","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9781139020411","volume-title":"Matrix analysis","author":"RA Horn","year":"2012","unstructured":"Horn RA, Johnson CR (2012) Matrix analysis. Cambridge University Press, Cambridge"},{"key":"1084_CR16","unstructured":"Hornik K, Meyer D, Schwendinger F, Theussl S (2020) ROI: R optimization infrastructure. R package version 1.0-0. https:\/\/CRAN.R-project.org\/package=ROI"},{"issue":"1","key":"1084_CR17","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1198\/0003130042836","volume":"58","author":"DR Hunter","year":"2004","unstructured":"Hunter DR, Lange K (2004) A tutorial on MM algorithms. Am Stat 58(1):30\u201337","journal-title":"Am Stat"},{"issue":"4","key":"1084_CR18","doi-asserted-by":"publisher","first-page":"373","DOI":"10.1007\/BF02579150","volume":"4","author":"N Karmarkar","year":"1984","unstructured":"Karmarkar N (1984) A new polynomial-time algorithm for linear programming. Combinatorica 4(4):373\u2013395. https:\/\/doi.org\/10.1007\/BF02579150","journal-title":"Combinatorica"},{"issue":"1","key":"1084_CR19","doi-asserted-by":"publisher","first-page":"291","DOI":"10.1007\/bf02613318","volume":"35","author":"H Kaufmann","year":"1988","unstructured":"Kaufmann H (1988) On existence and uniqueness of maximum likelihood estimates in quantal and ordinal response models. Metrika 35(1):291\u2013313. https:\/\/doi.org\/10.1007\/bf02613318","journal-title":"Metrika"},{"key":"1084_CR20","unstructured":"Konis K (2007) Linear programming algorithms for detecting separated data in binary logistic regression models. PhD thesis, University of Oxford. https:\/\/ora.ox.ac.uk\/objects\/uuid:8f9ee0d0-d78e-4101-9ab4-f9cbceed2a2a"},{"issue":"18","key":"1084_CR21","doi-asserted-by":"publisher","first-page":"1915","DOI":"10.1016\/j.spl.2009.05.020","volume":"79","author":"K Konis","year":"2009","unstructured":"Konis K, Fokianos K (2009) Safe density ratio modeling. Stat Probab Lett 79(18):1915\u20131920. https:\/\/doi.org\/10.1016\/j.spl.2009.05.020","journal-title":"Stat Probab Lett"},{"issue":"4","key":"1084_CR22","doi-asserted-by":"crossref","first-page":"392","DOI":"10.4310\/MAA.1994.v1.n4.a1","volume":"1","author":"K Lange","year":"1994","unstructured":"Lange K (1994) An adaptive barrier method for convex programming. Methods Appl Anal 1(4):392\u2013402","journal-title":"Methods Appl Anal"},{"issue":"1137\/1","key":"1084_CR23","first-page":"9781611974409","volume":"10","author":"K Lange","year":"2016","unstructured":"Lange K (2016) MM optimization algorithms. Soc Ind Appl Math 10(1137\/1):9781611974409","journal-title":"Soc Ind Appl Math"},{"key":"1084_CR24","unstructured":"Lumley T, Kronmal R, Ma S (2006) Relative risk regression in medical research: models, contrasts, estimators, and algorithms. Working Paper 293, UW Biostatistics Working Paper Series"},{"issue":"5","key":"1084_CR25","doi-asserted-by":"publisher","first-page":"981","DOI":"10.1007\/s00180-013-0476-8","volume":"29","author":"J Luo","year":"2014","unstructured":"Luo J, Zhang J, Sun H (2014) Estimation of relative risk using a log-binomial model with constraints. Comput Stat 29(5):981\u20131003. https:\/\/doi.org\/10.1007\/s00180-013-0476-8","journal-title":"Comput Stat"},{"key":"1084_CR26","unstructured":"Madsen K, Nielsen HB, Tingleff O (2004) Optimization with constraints, 2nd edn"},{"key":"1084_CR27","unstructured":"Makhorin A (2011) GNU linear programming kit reference manual version\u00a04.47. http:\/\/www.gnu.org\/software\/glpk"},{"key":"1084_CR28","doi-asserted-by":"publisher","first-page":"12","DOI":"10.32614\/RJ-2011-012","volume":"3","author":"IC Marschner","year":"2011","unstructured":"Marschner IC (2011) glm2: fitting generalized linear models with convergence problems. R J 3:12\u201315","journal-title":"R J"},{"issue":"4","key":"1084_CR29","doi-asserted-by":"publisher","first-page":"437","DOI":"10.1111\/anzs.12131","volume":"57","author":"IC Marschner","year":"2015","unstructured":"Marschner IC (2015) Relative risk regression for binary outcomes: methods and recommendations. Aust N Z J Stat 57(4):437\u2013462. https:\/\/doi.org\/10.1111\/anzs.12131","journal-title":"Aust N Z J Stat"},{"key":"1084_CR30","unstructured":"MOSEK ApS (2017) Introducing the MOSEK optimization suite. Version\u00a08.1 (Revision 27). http:\/\/docs.mosek.com\/8.1\/intro\/index.html"},{"key":"1084_CR31","doi-asserted-by":"publisher","unstructured":"Nemirovski A (2006) Advances in convex optimization: conic programming. In: Proceedings of international congress of mathematicians, pp 413\u2013444. https:\/\/doi.org\/10.4171\/022","DOI":"10.4171\/022"},{"key":"1084_CR32","doi-asserted-by":"crossref","unstructured":"Nesterov Y, Nemirovskii A (1994) Interior-point polynomial algorithms in convex programming. SIAM Studies in Applied Mathematics, Society for Industrial and Applied Mathematics, Philadelphia","DOI":"10.1137\/1.9781611970791"},{"key":"1084_CR33","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-40065-5","volume-title":"Numerical optimization","author":"J Nocedal","year":"2006","unstructured":"Nocedal J, Wright SJ (2006) Numerical optimization. Springer, Berlin. https:\/\/doi.org\/10.1007\/978-0-387-40065-5"},{"key":"1084_CR34","unstructured":"O\u2019Donoghue B (2015) SCS\u2013(splitting conic solver). https:\/\/github.com\/cvxgrp\/scs.git"},{"key":"1084_CR35","doi-asserted-by":"publisher","unstructured":"O\u2019Donoghue B, Chu E, Parikh N, Boyd S (2016) Conic optimization via operator splitting and homogeneous self-dual embedding. J Optim Theory Appl 1\u201327. https:\/\/doi.org\/10.1007\/s10957-016-0892-3","DOI":"10.1007\/s10957-016-0892-3"},{"key":"1084_CR36","unstructured":"Ortega JM, Rheinboldt WC (1970) Iterative solution of nonlinear equations in several variables, vol 30. Society for Industrial and Applied Mathematics"},{"key":"1084_CR37","unstructured":"R Core Team (2020) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https:\/\/www.R-project.org\/"},{"issue":"22","key":"1084_CR38","doi-asserted-by":"publisher","first-page":"2269","DOI":"10.1002\/sim.3989","volume":"29","author":"A Savu","year":"2010","unstructured":"Savu A, Liu Q, Yasui Y (2010) Estimation of relative risk and prevalence ratio. Stat Med 29(22):2269\u20132281. https:\/\/doi.org\/10.1002\/sim.3989","journal-title":"Stat Med"},{"key":"1084_CR39","unstructured":"Serrano SA (2015) Algorithms for unsymmetric cone optimization and an implementation for problems with the exponential cone. PhD thesis, Stanford University. https:\/\/web.stanford.edu\/group\/SOL\/dissertations\/ThesisAkleAdobe-augmented.pdf"},{"issue":"3","key":"1084_CR40","doi-asserted-by":"crossref","first-page":"310","DOI":"10.1111\/j.2517-6161.1981.tb01676.x","volume":"43","author":"MJ Silvapulle","year":"1981","unstructured":"Silvapulle MJ (1981) On the existence of maximum likelihood estimators for the binomial response models. J Roy Stat Soc Ser B (Methodol) 43(3):310\u2013313","journal-title":"J Roy Stat Soc Ser B (Methodol)"},{"issue":"15","key":"1084_CR41","doi-asserted-by":"publisher","first-page":"1","DOI":"10.18637\/jss.v094.i15","volume":"94","author":"S Theu\u00dfl","year":"2020","unstructured":"Theu\u00dfl S, Schwendinger F, Hornik K (2020) ROI: an extensible R optimization infrastructure. J Stat Softw 94(15):1\u201364. https:\/\/doi.org\/10.18637\/jss.v094.i15","journal-title":"J Stat Softw"},{"key":"1084_CR42","unstructured":"Varadhan R (2015) alabama: constrained nonlinear optimization. R package version 2015.3-1. https:\/\/CRAN.R-project.org\/package=alabama"},{"key":"1084_CR43","unstructured":"W\u00e4chter A (2009) Short tutorial: Getting started with Ipopt in 90 minutes. In: Naumann U, Schenk O, Simon HD, Toledo S (eds) Combinatorial scientific computing, Schloss Dagstuhl\u2013Leibniz-Zentrum f\u00fcr Informatik, Germany, Dagstuhl, Germany, no. 09061 in Dagstuhl Seminar Proceedings. http:\/\/drops.dagstuhl.de\/opus\/volltexte\/2009\/2089"},{"issue":"1","key":"1084_CR44","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1007\/s10107-004-0559-y","volume":"106","author":"A W\u00e4chter","year":"2006","unstructured":"W\u00e4chter A, Biegler LT (2006) On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming. Math Program 106(1):25\u201357. https:\/\/doi.org\/10.1007\/s10107-004-0559-y","journal-title":"Math Program"},{"key":"1084_CR45","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1186\/1742-7622-10-14","volume":"10","author":"T Williamson","year":"2013","unstructured":"Williamson T, Eliasziw M, Fick GH (2013) Log-binomial models: exploring failed convergence. Emerg Themes Epidemiol 10:14. https:\/\/doi.org\/10.1186\/1742-7622-10-14","journal-title":"Emerg Themes Epidemiol"}],"container-title":["Computational Statistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00180-021-01084-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00180-021-01084-5\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00180-021-01084-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,24]],"date-time":"2024-08-24T16:52:48Z","timestamp":1724518368000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00180-021-01084-5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,2,22]]},"references-count":45,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2021,9]]}},"alternative-id":["1084"],"URL":"https:\/\/doi.org\/10.1007\/s00180-021-01084-5","relation":{},"ISSN":["0943-4062","1613-9658"],"issn-type":[{"value":"0943-4062","type":"print"},{"value":"1613-9658","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,2,22]]},"assertion":[{"value":"25 February 2020","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 January 2021","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 February 2021","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}