{"created":"2023-07-27T06:19:34.865641+00:00","id":459,"links":{},"metadata":{"_buckets":{"deposit":"a5308c07-7d87-4661-86df-1eb407f972dd"},"_deposit":{"created_by":3,"id":"459","owners":[3],"pid":{"revision_id":0,"type":"depid","value":"459"},"status":"published"},"_oai":{"id":"oai:kanazawa-u.repo.nii.ac.jp:00000459","sets":["11:12:15"]},"author_link":["3183","3184","331","86010"],"item_4_biblio_info_8":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"2010-03-01","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"2","bibliographicPageEnd":"298","bibliographicPageStart":"265","bibliographicVolumeNumber":"39","bibliographic_titles":[{"bibliographic_title":"日本統計学会誌. シリーズJ = Journal of the Japan Statistical Society. Japanese issue"}]}]},"item_4_creator_33":{"attribute_name":"著者別表示","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Sagae, Masahiko"}],"nameIdentifiers":[{},{}]}]},"item_4_description_21":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"統計処理の対象となるデータは近年,大規模化,多次元化の方向へ急激に増大しつつある.ノンパラメトリック密度推定の研究は,カーネル密度推定の理論が整備され,実用的な面では主な統計ソフトウェアに標準装備され,容易に利用できる.しかし大規模データと多次元データに対応したノンパラメトリック統計解析の研究は進んでいない.この原因は,多次元カーネル関数の構成法が未開発であることと,カーネル推定では大規模データの計算負荷を大幅に改善する方法は今のところなく,計算機のハード及びソフトのパワーに多くを依存していることによると考えられる.そこで我々は大規模データへの計算負荷を軽減する方法として,ビン型推定に着目する.多次元データの推定精度の問題はあまり議論されていないが,多次元密度推定において次元が増加するにつれて推定精度は低下する.したがって,カーネル推定でこれを改善するには多次元かつ高次オーダー特性(収束率)をもつカーネル関数の利用が必要となるが,この種のカーネル関数は現在のところ知られていない.ヒストグラムに代表されるビン型推定の計算負荷は相対的に少ないが,推定精度の低下を伴うことがよく知られている.近年,推定精度を保つビン型推定としてPolynomial Histogram(Sagae and Scott(1997))が提案され,計算負荷を軽減しつつ,カーネル推定と同等な推定精度を持つことが示された.本稿では多次元データへこの方法を拡張し,多次元Polynomialヒストグラムを提案する.この密度関数は高次オーダー特性を持ちつつ,大規模データに対する計算負荷の低減が可能であることを示す.また,数値例でその有効性を示す. This paper extends polynomial histogram Sagae and Scott (1997) to multi-dimensional case. There are a number of recent studies about the multi-dimensional kernel density estimation. However, the multi-dimensional kernel density estimation with higher-order asymptotic performance is undeveloped as far as I know and also studies of the multi-dimensional density function have been stagnant on grouping data. We propose a multi-dimensional polynomial histogram and prove higher-order asymptotic property. Numerical experiments confirm the effectiveness.","subitem_description_type":"Abstract"}]},"item_4_identifier_registration":{"attribute_name":"ID登録","attribute_value_mlt":[{"subitem_identifier_reg_text":"10.24517/00000449","subitem_identifier_reg_type":"JaLC"}]},"item_4_publisher_17":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"日本統計学会 = Japan Statistical Society"}]},"item_4_relation_28":{"attribute_name":"関連URI","attribute_value_mlt":[{"subitem_relation_type_id":{"subitem_relation_type_id_text":"http://www.jss.gr.jp/index.html","subitem_relation_type_select":"URI"}},{"subitem_relation_type_id":{"subitem_relation_type_id_text":"http://ci.nii.ac.jp/naid/110007618317","subitem_relation_type_select":"URI"}}]},"item_4_rights_23":{"attribute_name":"権利","attribute_value_mlt":[{"subitem_rights":"日本統計学会 | 本文データは学協会の許諾に基づきCiNiiから複製したものである"}]},"item_4_source_id_11":{"attribute_name":"NCID","attribute_value_mlt":[{"subitem_source_identifier":"AA11989749","subitem_source_identifier_type":"NCID"}]},"item_4_source_id_9":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"0389-5602","subitem_source_identifier_type":"ISSN"}]},"item_4_text_2":{"attribute_name":"その他のタイトル","attribute_value_mlt":[{"subitem_text_value":"<特集>ノンパラメトリック法の新展開"},{"subitem_text_value":"Multivariate Polynomial Histogram Density Estimation(<Special Section>New Developments in Nonparametric Method)"}]},"item_4_version_type_25":{"attribute_name":"著者版フラグ","attribute_value_mlt":[{"subitem_version_resource":"http://purl.org/coar/version/c_970fb48d4fbd8a85","subitem_version_type":"VoR"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"寒河江, 雅彦"}],"nameIdentifiers":[{},{},{},{}]},{"creatorNames":[{"creatorName":"野呂, 拓矢"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"スコット デビッド"}],"nameIdentifiers":[{}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2017-10-02"}],"displaytype":"detail","filename":"EC-PR-SAGAE-M-265.pdf","filesize":[{"value":"1.6 MB"}],"format":"application/pdf","licensetype":"license_11","mimetype":"application/pdf","url":{"label":"EC-PR-SAGAE-M-265.pdf","url":"https://kanazawa-u.repo.nii.ac.jp/record/459/files/EC-PR-SAGAE-M-265.pdf"},"version_id":"64ce5cab-186f-4510-8d3a-8cf19370f607"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourcetype":"journal article","resourceuri":"http://purl.org/coar/resource_type/c_6501"}]},"item_title":"Polynomial Histogramによる多次元ノンパラメトリック確率密度推定","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"Polynomial Histogramによる多次元ノンパラメトリック確率密度推定"}]},"item_type_id":"4","owner":"3","path":["15"],"pubdate":{"attribute_name":"公開日","attribute_value":"2017-10-02"},"publish_date":"2017-10-02","publish_status":"0","recid":"459","relation_version_is_last":true,"title":["Polynomial Histogramによる多次元ノンパラメトリック確率密度推定"],"weko_creator_id":"3","weko_shared_id":3},"updated":"2023-07-27T08:49:17.066217+00:00"}