Estimation in the convolution structure density model. Part I: oracle inequalities

We study the problem of nonparametric estimation under $\bL_p$-loss, $p\in [1,\infty)$, in the framework of the convolution structure density model on $\bR^d$. This observation scheme is a generalization of two classical statistical models, namely density estimation under direct and indirect observa...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Lepski, Oleg, Willer, Thomas
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Lepski, Oleg
Willer, Thomas
description We study the problem of nonparametric estimation under $\bL_p$-loss, $p\in [1,\infty)$, in the framework of the convolution structure density model on $\bR^d$. This observation scheme is a generalization of two classical statistical models, namely density estimation under direct and indirect observations. In Part I the original pointwise selection rule from a family of "kernel-type" estimators is proposed. For the selected estimator, we prove an $\bL_p$-norm oracle inequality and several of its consequences. In Part II the problem of adaptive minimax estimation under $\bL_p$--loss over the scale of anisotropic Nikol'skii classes is addressed. We fully characterize the behavior of the minimax risk for different relationships between regularity parameters and norm indexes in the definitions of the functional class and of the risk. We prove that the selection rule proposed in Part I leads to the construction of an optimally or nearly optimally (up to logarithmic factor) adaptive estimator.
doi_str_mv 10.48550/arxiv.1704.04418
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1704_04418</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1704_04418</sourcerecordid><originalsourceid>FETCH-LOGICAL-a678-2000497560eb613e9588928cbfcf3ac9b1fb6c0bc6f94dd26300d3286d2e37423</originalsourceid><addsrcrecordid>eNotz99KwzAYBfDceCHTB_DKvEDr1yRNE-9kTB0MJrL7kj9fMNA1mqbDvb2zenXgwDnwI-SugVqotoUHk7_jqW46EDUI0ahr8r6ZSjyaEtNI40jLB1KXxlMa5qWaSp5dmTNSj-MUy5kek8ehpm8mF7p9pCkbN-Blil-zGWKJON2Qq2CGCW__c0UOz5vD-rXa7V-266ddZWSnKgYAQnetBLSy4ahbpTRTzgYXuHHaNsFKB9bJoIX3THIAz5mSniHvBOMrcv93u6D6z3xh5HP_i-sXHP8BQslKeg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Estimation in the convolution structure density model. Part I: oracle inequalities</title><source>arXiv.org</source><creator>Lepski, Oleg ; Willer, Thomas</creator><creatorcontrib>Lepski, Oleg ; Willer, Thomas</creatorcontrib><description>We study the problem of nonparametric estimation under $\bL_p$-loss, $p\in [1,\infty)$, in the framework of the convolution structure density model on $\bR^d$. This observation scheme is a generalization of two classical statistical models, namely density estimation under direct and indirect observations. In Part I the original pointwise selection rule from a family of "kernel-type" estimators is proposed. For the selected estimator, we prove an $\bL_p$-norm oracle inequality and several of its consequences. In Part II the problem of adaptive minimax estimation under $\bL_p$--loss over the scale of anisotropic Nikol'skii classes is addressed. We fully characterize the behavior of the minimax risk for different relationships between regularity parameters and norm indexes in the definitions of the functional class and of the risk. We prove that the selection rule proposed in Part I leads to the construction of an optimally or nearly optimally (up to logarithmic factor) adaptive estimator.</description><identifier>DOI: 10.48550/arxiv.1704.04418</identifier><language>eng</language><subject>Mathematics - Statistics Theory ; Statistics - Theory</subject><creationdate>2017-04</creationdate><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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,777,882</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1704.04418$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1704.04418$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Lepski, Oleg</creatorcontrib><creatorcontrib>Willer, Thomas</creatorcontrib><title>Estimation in the convolution structure density model. Part I: oracle inequalities</title><description>We study the problem of nonparametric estimation under $\bL_p$-loss, $p\in [1,\infty)$, in the framework of the convolution structure density model on $\bR^d$. This observation scheme is a generalization of two classical statistical models, namely density estimation under direct and indirect observations. In Part I the original pointwise selection rule from a family of "kernel-type" estimators is proposed. For the selected estimator, we prove an $\bL_p$-norm oracle inequality and several of its consequences. In Part II the problem of adaptive minimax estimation under $\bL_p$--loss over the scale of anisotropic Nikol'skii classes is addressed. We fully characterize the behavior of the minimax risk for different relationships between regularity parameters and norm indexes in the definitions of the functional class and of the risk. We prove that the selection rule proposed in Part I leads to the construction of an optimally or nearly optimally (up to logarithmic factor) adaptive estimator.</description><subject>Mathematics - Statistics Theory</subject><subject>Statistics - Theory</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2017</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz99KwzAYBfDceCHTB_DKvEDr1yRNE-9kTB0MJrL7kj9fMNA1mqbDvb2zenXgwDnwI-SugVqotoUHk7_jqW46EDUI0ahr8r6ZSjyaEtNI40jLB1KXxlMa5qWaSp5dmTNSj-MUy5kek8ehpm8mF7p9pCkbN-Blil-zGWKJON2Qq2CGCW__c0UOz5vD-rXa7V-266ddZWSnKgYAQnetBLSy4ahbpTRTzgYXuHHaNsFKB9bJoIX3THIAz5mSniHvBOMrcv93u6D6z3xh5HP_i-sXHP8BQslKeg</recordid><startdate>20170414</startdate><enddate>20170414</enddate><creator>Lepski, Oleg</creator><creator>Willer, Thomas</creator><scope>AKZ</scope><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20170414</creationdate><title>Estimation in the convolution structure density model. Part I: oracle inequalities</title><author>Lepski, Oleg ; Willer, Thomas</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a678-2000497560eb613e9588928cbfcf3ac9b1fb6c0bc6f94dd26300d3286d2e37423</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2017</creationdate><topic>Mathematics - Statistics Theory</topic><topic>Statistics - Theory</topic><toplevel>online_resources</toplevel><creatorcontrib>Lepski, Oleg</creatorcontrib><creatorcontrib>Willer, Thomas</creatorcontrib><collection>arXiv Mathematics</collection><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Lepski, Oleg</au><au>Willer, Thomas</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Estimation in the convolution structure density model. Part I: oracle inequalities</atitle><date>2017-04-14</date><risdate>2017</risdate><abstract>We study the problem of nonparametric estimation under $\bL_p$-loss, $p\in [1,\infty)$, in the framework of the convolution structure density model on $\bR^d$. This observation scheme is a generalization of two classical statistical models, namely density estimation under direct and indirect observations. In Part I the original pointwise selection rule from a family of "kernel-type" estimators is proposed. For the selected estimator, we prove an $\bL_p$-norm oracle inequality and several of its consequences. In Part II the problem of adaptive minimax estimation under $\bL_p$--loss over the scale of anisotropic Nikol'skii classes is addressed. We fully characterize the behavior of the minimax risk for different relationships between regularity parameters and norm indexes in the definitions of the functional class and of the risk. We prove that the selection rule proposed in Part I leads to the construction of an optimally or nearly optimally (up to logarithmic factor) adaptive estimator.</abstract><doi>10.48550/arxiv.1704.04418</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1704.04418
ispartof
issn
language eng
recordid cdi_arxiv_primary_1704_04418
source arXiv.org
subjects Mathematics - Statistics Theory
Statistics - Theory
title Estimation in the convolution structure density model. Part I: oracle inequalities
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-18T15%3A30%3A36IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Estimation%20in%20the%20convolution%20structure%20density%20model.%20Part%20I:%20oracle%20inequalities&rft.au=Lepski,%20Oleg&rft.date=2017-04-14&rft_id=info:doi/10.48550/arxiv.1704.04418&rft_dat=%3Carxiv_GOX%3E1704_04418%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true