Fast and Accurate Estimation of Non-Nested Binomial Hierarchical Models Using Variational Inference

Non-linear hierarchical models are commonly used in many disciplines. However, inference in the presence of non-nested effects and on large datasets is challenging and computationally burdensome. This paper provides two contributions to scalable and accurate inference. First, I derive a new mean-fie...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2021-10
1. Verfasser: Goplerud, Max
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Goplerud, Max
description Non-linear hierarchical models are commonly used in many disciplines. However, inference in the presence of non-nested effects and on large datasets is challenging and computationally burdensome. This paper provides two contributions to scalable and accurate inference. First, I derive a new mean-field variational algorithm for estimating binomial logistic hierarchical models with an arbitrary number of non-nested random effects. Second, I propose "marginally augmented variational Bayes" (MAVB) that further improves the initial approximation through a step of Bayesian post-processing. I prove that MAVB provides a guaranteed improvement in the approximation quality at low computational cost and induces dependencies that were assumed away by the initial factorization assumptions. I apply these techniques to a study of voter behavior using a high-dimensional application of the popular approach of multilevel regression and post-stratification (MRP). Existing estimation took hours whereas the algorithms proposed run in minutes. The posterior means are well-recovered even under strong factorization assumptions. Applying MAVB further improves the approximation by partially correcting the under-estimated variance. The proposed methodology is implemented in an open source software package.
doi_str_mv 10.48550/arxiv.2007.12300
format Article
fullrecord <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2007_12300</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2427408076</sourcerecordid><originalsourceid>FETCH-LOGICAL-a526-b70fdca7aaeedd9af6dde3d1f23fd59b7f001baddd9337196dda179d84e4f9f03</originalsourceid><addsrcrecordid>eNotkE9PAjEQxRsTEwnyATzZxPPitN3d7h6RgJAgXtDrZugfLYEW28Xot7eip8nk995k3iPkhsG4bKoK7jF-uc8xB5BjxgXABRlwIVjRlJxfkVFKOwDgteRVJQZEzTH1FL2mE6VOEXtDZ6l3B-xd8DRYug6-WJvUG00fnA8Hh3u6cCZiVO9O5eUpaLNP9CU5_0ZfMbqzNYOltyYar8w1ubS4T2b0P4dkM59tpoti9fy4nE5WBVa8LrYSrFYoEY3RukVba22EZpYLq6t2Ky0A26LOTAjJ2oyRyVY3pSlta0EMye3f2XMD3THmFPG7-22iOzeRFXd_imMMH6ccqtuFU8y_po6XXJbQgKzFD9o3YuU</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2427408076</pqid></control><display><type>article</type><title>Fast and Accurate Estimation of Non-Nested Binomial Hierarchical Models Using Variational Inference</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Goplerud, Max</creator><creatorcontrib>Goplerud, Max</creatorcontrib><description>Non-linear hierarchical models are commonly used in many disciplines. However, inference in the presence of non-nested effects and on large datasets is challenging and computationally burdensome. This paper provides two contributions to scalable and accurate inference. First, I derive a new mean-field variational algorithm for estimating binomial logistic hierarchical models with an arbitrary number of non-nested random effects. Second, I propose "marginally augmented variational Bayes" (MAVB) that further improves the initial approximation through a step of Bayesian post-processing. I prove that MAVB provides a guaranteed improvement in the approximation quality at low computational cost and induces dependencies that were assumed away by the initial factorization assumptions. I apply these techniques to a study of voter behavior using a high-dimensional application of the popular approach of multilevel regression and post-stratification (MRP). Existing estimation took hours whereas the algorithms proposed run in minutes. The posterior means are well-recovered even under strong factorization assumptions. Applying MAVB further improves the approximation by partially correcting the under-estimated variance. The proposed methodology is implemented in an open source software package.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2007.12300</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Approximation ; Estimation ; Factorization ; Inference ; Mathematical analysis ; Post-production processing ; Statistics - Methodology</subject><ispartof>arXiv.org, 2021-10</ispartof><rights>2021. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,780,881,27904</link.rule.ids><backlink>$$Uhttps://doi.org/10.1214/21-BA1266$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.48550/arXiv.2007.12300$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Goplerud, Max</creatorcontrib><title>Fast and Accurate Estimation of Non-Nested Binomial Hierarchical Models Using Variational Inference</title><title>arXiv.org</title><description>Non-linear hierarchical models are commonly used in many disciplines. However, inference in the presence of non-nested effects and on large datasets is challenging and computationally burdensome. This paper provides two contributions to scalable and accurate inference. First, I derive a new mean-field variational algorithm for estimating binomial logistic hierarchical models with an arbitrary number of non-nested random effects. Second, I propose "marginally augmented variational Bayes" (MAVB) that further improves the initial approximation through a step of Bayesian post-processing. I prove that MAVB provides a guaranteed improvement in the approximation quality at low computational cost and induces dependencies that were assumed away by the initial factorization assumptions. I apply these techniques to a study of voter behavior using a high-dimensional application of the popular approach of multilevel regression and post-stratification (MRP). Existing estimation took hours whereas the algorithms proposed run in minutes. The posterior means are well-recovered even under strong factorization assumptions. Applying MAVB further improves the approximation by partially correcting the under-estimated variance. The proposed methodology is implemented in an open source software package.</description><subject>Algorithms</subject><subject>Approximation</subject><subject>Estimation</subject><subject>Factorization</subject><subject>Inference</subject><subject>Mathematical analysis</subject><subject>Post-production processing</subject><subject>Statistics - Methodology</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNotkE9PAjEQxRsTEwnyATzZxPPitN3d7h6RgJAgXtDrZugfLYEW28Xot7eip8nk995k3iPkhsG4bKoK7jF-uc8xB5BjxgXABRlwIVjRlJxfkVFKOwDgteRVJQZEzTH1FL2mE6VOEXtDZ6l3B-xd8DRYug6-WJvUG00fnA8Hh3u6cCZiVO9O5eUpaLNP9CU5_0ZfMbqzNYOltyYar8w1ubS4T2b0P4dkM59tpoti9fy4nE5WBVa8LrYSrFYoEY3RukVba22EZpYLq6t2Ky0A26LOTAjJ2oyRyVY3pSlta0EMye3f2XMD3THmFPG7-22iOzeRFXd_imMMH6ccqtuFU8y_po6XXJbQgKzFD9o3YuU</recordid><startdate>20211020</startdate><enddate>20211020</enddate><creator>Goplerud, Max</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20211020</creationdate><title>Fast and Accurate Estimation of Non-Nested Binomial Hierarchical Models Using Variational Inference</title><author>Goplerud, Max</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a526-b70fdca7aaeedd9af6dde3d1f23fd59b7f001baddd9337196dda179d84e4f9f03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Algorithms</topic><topic>Approximation</topic><topic>Estimation</topic><topic>Factorization</topic><topic>Inference</topic><topic>Mathematical analysis</topic><topic>Post-production processing</topic><topic>Statistics - Methodology</topic><toplevel>online_resources</toplevel><creatorcontrib>Goplerud, Max</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Goplerud, Max</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Fast and Accurate Estimation of Non-Nested Binomial Hierarchical Models Using Variational Inference</atitle><jtitle>arXiv.org</jtitle><date>2021-10-20</date><risdate>2021</risdate><eissn>2331-8422</eissn><abstract>Non-linear hierarchical models are commonly used in many disciplines. However, inference in the presence of non-nested effects and on large datasets is challenging and computationally burdensome. This paper provides two contributions to scalable and accurate inference. First, I derive a new mean-field variational algorithm for estimating binomial logistic hierarchical models with an arbitrary number of non-nested random effects. Second, I propose "marginally augmented variational Bayes" (MAVB) that further improves the initial approximation through a step of Bayesian post-processing. I prove that MAVB provides a guaranteed improvement in the approximation quality at low computational cost and induces dependencies that were assumed away by the initial factorization assumptions. I apply these techniques to a study of voter behavior using a high-dimensional application of the popular approach of multilevel regression and post-stratification (MRP). Existing estimation took hours whereas the algorithms proposed run in minutes. The posterior means are well-recovered even under strong factorization assumptions. Applying MAVB further improves the approximation by partially correcting the under-estimated variance. The proposed methodology is implemented in an open source software package.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2007.12300</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2021-10
issn 2331-8422
language eng
recordid cdi_arxiv_primary_2007_12300
source arXiv.org; Free E- Journals
subjects Algorithms
Approximation
Estimation
Factorization
Inference
Mathematical analysis
Post-production processing
Statistics - Methodology
title Fast and Accurate Estimation of Non-Nested Binomial Hierarchical Models Using Variational Inference
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-26T22%3A25%3A34IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Fast%20and%20Accurate%20Estimation%20of%20Non-Nested%20Binomial%20Hierarchical%20Models%20Using%20Variational%20Inference&rft.jtitle=arXiv.org&rft.au=Goplerud,%20Max&rft.date=2021-10-20&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2007.12300&rft_dat=%3Cproquest_arxiv%3E2427408076%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2427408076&rft_id=info:pmid/&rfr_iscdi=true