{"id":123984,"date":"2026-04-20T12:21:49","date_gmt":"2026-04-20T10:21:49","guid":{"rendered":"https:\/\/www.istat.it\/?page_id=123984"},"modified":"2026-04-20T12:41:30","modified_gmt":"2026-04-20T10:41:30","slug":"univoutl","status":"publish","type":"page","link":"https:\/\/www.istat.it\/en\/classifications-and-tools\/methods-and-software-of-the-statistical-process\/process-phase\/detection-and-treatment-of-measurement-errors-and-imputation-of-partial-non-responses\/univoutl\/","title":{"rendered":"UnivOutl"},"content":{"rendered":"\n<p>The contents related to univOutl are shown in the following sections:<\/p>\n\n\n<section class=\"gblock accordion_livelli white-bg  py-0\"  aria-labelledby=\"section-1\"><div class=\"container p-lg-0 block_count_1\" data-blockcount=\"1\"><div class=\"row pb-2\">\t\t<div class=\"accordion bianco\" id=\"accordion_1liv_0_block_70440fae73a86b2168c64e7a83ac57d6\">\r\n\t\t\t\t\t\t\t<div class=\"accordion-item liv1\">\r\n\t\t\t\t\t\t\t\t\t\t\t<h2 class=\"accordion-header\">\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<button class=\"accordion-button collapsed\" type=\"button\" \r\n\t\t\t\t\t\t\t\tdata-bs-toggle=\"collapse\" data-bs-target=\"#collapse_al_1_1liv_block_70440fae73a86b2168c64e7a83ac57d6\" aria-expanded=\"false\" aria-controls=\"collapse_al_1_1liv_block_70440fae73a86b2168c64e7a83ac57d6\">\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tDescription\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/button>\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t<\/h2>\r\n\t\t\t\t\t\t\t\t\t\t<div id=\"collapse_al_1_1liv_block_70440fae73a86b2168c64e7a83ac57d6\" class=\"accordion-collapse collapse\" \r\n\t\t\t\t\t\tdata-row=\"0\" \r\n\t\t\t\t\t\tdata-depth=\"1\" \r\n\t\t\t\t\t\t\t\t\t\t\t\tdata-bs-parent=\"#accordion_1liv_0_block_70440fae73a86b2168c64e7a83ac57d6\">\r\n                        <div class=\"row\">\r\n                                                            <div class=\"accordion-body col-12\">\r\n                                    <p>The <strong>univOutl<\/strong> package is package for the R environment that implements the main techniques for identifying outliers in data related to a single quantitative variable (univariate outliers) or in cases where the same quantitative variable is observed on the same units in different occasions, such as in panel surveys.<\/p>\n<p>The methods for identifying univariate outliers are essentially based on two approaches:<\/p>\n<ul class=\"lista\">\n<li>The assumption of a Gaussian model for the data distribution;<\/li>\n<li>A non-parametric approach based on the use of boxplots.<\/li>\n<\/ul>\n<p>When a Gaussian distribution is assumed, the univOutl package offers various solutions for the robust estimation of distribution parameters (mean and standard deviation) to reduce the influence of extreme values.<\/p>\n<p>The non-parametric approach, on the other hand, is more flexible as it does not require a specific distributional model; it only requires an evaluation of the degree of symmetry in the data distribution. With symmetric distributions the outliers are identified using the traditional boxplot. In presence of asymmetric distributions, a common situation for variables related to businesses, farms, or household economic data, methods based on boxplots are appropriately modified to account for skewness.<\/p>\n<p>Finally, the package includes specific tools for identifying outliers in continuous variables observed on the same units at two different points in time. These methods are based on the construction of ratios, including the Hidiroglou-Berthelot (1986) method and a non-parametric extension of it which introduces an additional degree of flexibility.<\/p>\n<p>It should also be noted that, within the context of sample surveys, several functions in the univOutl package allow for the incorporation of survey weights into the analysis.<\/p>\n<p>&nbsp;<\/p>\n<p>Main references<\/p>\n<p>Hidiroglou, M.A. and Berthelot, J.-M. (1986) \u201cStatistical editing and Imputation for Periodic Business Surveys\u201d. <em>Survey Methodology<\/em>, Vol 12, pp. 73-83.<\/p>\n<p>McGill, R., Tukey, J. W. and Larsen, W. A. (1978) \u201cVariations of box plots\u201d. <em>The American Statistician<\/em>, 32, pp. 12-16.<\/p>\n<p>Rousseeuw, P.J. and Croux, C. (1993) \u201cAlternatives to the Median Absolute Deviation\u201d, <em>Journal of the American Statistical Association<\/em>, 88, pp. 1273-1283.<\/p>\n<p>Hubert, M., and Vandervieren, E. (2008) \u201cAn Adjusted Boxplot for Skewed Distributions\u201d, <em>Computational Statistics &amp; Data Analysis<\/em>, 52, pp. 5186-5201<\/p>\n                                <\/div>\r\n                                                                                <\/div>\r\n                        \t\t\t\t\t<\/div>\r\n\t\t\t\t<\/div>\r\n\t\t\t\t\t\t\t<div class=\"accordion-item liv1\">\r\n\t\t\t\t\t\t\t\t\t\t\t<h2 class=\"accordion-header\">\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<button class=\"accordion-button collapsed\" type=\"button\" \r\n\t\t\t\t\t\t\t\tdata-bs-toggle=\"collapse\" data-bs-target=\"#collapse_al_2_1liv_block_70440fae73a86b2168c64e7a83ac57d6\" aria-expanded=\"false\" aria-controls=\"collapse_al_2_1liv_block_70440fae73a86b2168c64e7a83ac57d6\">\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tInformation\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/button>\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t<\/h2>\r\n\t\t\t\t\t\t\t\t\t\t<div id=\"collapse_al_2_1liv_block_70440fae73a86b2168c64e7a83ac57d6\" class=\"accordion-collapse collapse\" \r\n\t\t\t\t\t\tdata-row=\"0\" \r\n\t\t\t\t\t\tdata-depth=\"1\" \r\n\t\t\t\t\t\t\t\t\t\t\t\tdata-bs-parent=\"#accordion_1liv_0_block_70440fae73a86b2168c64e7a83ac57d6\">\r\n                        <div class=\"row\">\r\n                                                            <div class=\"accordion-body col-12\">\r\n                                    <p><strong>Status:<\/strong> validated<\/p>\n<p><strong>Author:<\/strong> Istat<\/p>\n<p><strong>Licence:<\/strong> GPL-2 | GPL-3<\/p>\n<p><strong>GSBPM code:<\/strong>5.4 Edit and impute<\/p>\n<p><strong>Programming language:<\/strong> R<\/p>\n<p><strong>Keywords: <\/strong>robust estimation; boxplot; Hidiroglou-Berthelot method<\/p>\n<p><strong>Contact:\u00a0<\/strong>name: Marcello D\u2019Orazio &#8211; email: <a href=\"mailto:madorazi@istat.it\">madorazi@istat.it<\/a><\/p>\n                                <\/div>\r\n                                                                                <\/div>\r\n                        \t\t\t\t\t<\/div>\r\n\t\t\t\t<\/div>\r\n\t\t\t\t\t\t\t<div class=\"accordion-item liv1\">\r\n\t\t\t\t\t\t\t\t\t\t\t<h2 class=\"accordion-header\">\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<button class=\"accordion-button collapsed\" type=\"button\" \r\n\t\t\t\t\t\t\t\tdata-bs-toggle=\"collapse\" data-bs-target=\"#collapse_al_3_1liv_block_70440fae73a86b2168c64e7a83ac57d6\" aria-expanded=\"false\" aria-controls=\"collapse_al_3_1liv_block_70440fae73a86b2168c64e7a83ac57d6\">\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tSoftware and documentation\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/button>\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t<\/h2>\r\n\t\t\t\t\t\t\t\t\t\t<div id=\"collapse_al_3_1liv_block_70440fae73a86b2168c64e7a83ac57d6\" class=\"accordion-collapse collapse\" \r\n\t\t\t\t\t\tdata-row=\"0\" \r\n\t\t\t\t\t\tdata-depth=\"1\" \r\n\t\t\t\t\t\t\t\t\t\t\t\tdata-bs-parent=\"#accordion_1liv_0_block_70440fae73a86b2168c64e7a83ac57d6\">\r\n                        <div class=\"row\">\r\n                                                            <div class=\"accordion-body col-12\">\r\n                                    <p><strong>TECHNICAL REQUIREMENTS<\/strong><\/p>\n<p>The univOutl package works on R versions 3.6.0 and later on any operating system (Windows, Mac, or Linux). It requires the following additional R packages to be installed and subsequently loaded: robustbase and Hmisc.<\/p>\n<p>&nbsp;<\/p>\n<p><strong>COPYRIGHT<\/strong><\/p>\n<p>Copyright 2026 Marcello D\u2019Orazio<\/p>\n<p>Licensed under the GNU General Public License (GPL), version 2 or later. You may not use this work except in compliance with the License. You may obtain a copy of the License at the following address: http:\/\/www.gnu.org\/licenses\/.<\/p>\n<p>Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an &#8220;AS IS&#8221; BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.<\/p>\n<p>&nbsp;<\/p>\n<p><strong>DISCLAIMER<\/strong><\/p>\n<p>Istat does not assume responsibility for results deriving from a use of the tool that is not consistent with the methodological indications contained in the available documentation.<\/p>\n<p>&nbsp;<\/p>\n<p><strong>DOWNLOAD<\/strong><\/p>\n<p>Release date: 21\/02\/2026<\/p>\n<p><a href=\"https:\/\/www.istat.it\/wp-content\/uploads\/2026\/04\/univOutl_0.5.0.zip\">univOutl Version 0.5.0<\/a> \u2013 Precompiled package for Windows<\/p>\n<p><a href=\"https:\/\/www.istat.it\/wp-content\/uploads\/2026\/04\/univOutl_0.5.0.tar.gz\">univOutl Version 0.5.0<\/a> \u2013 Package source for Windows and Unix-like systems<\/p>\n<p>&nbsp;<\/p>\n<p><strong>INSTALLATION<\/strong><\/p>\n<p>In R the package can be installed using the following instructions:<\/p>\n<p>&gt; install.packages(path_to_file, repos = NULL)<\/p>\n<p>where path_to_file indicated the path of the downloaded.zip or .tar.gz.<\/p>\n<p>&nbsp;<\/p>\n<p><strong>TECHNICAL AND METHODOLOGICAL DOCUMENTATION<\/strong><\/p>\n<p>Reference manual \u2013 univOutl v. 0.5.0<\/p>\n<p><a href=\"https:\/\/cran.r-project.org\/web\/packages\/univOutl\/univOutl.pdf\">https:\/\/cran.r-project.org\/web\/packages\/univOutl\/univOutl.pdf<\/a><\/p>\n<p>Hidiroglou, M.A. and Berthelot, J.-M. (1986) \u201cStatistical editing and Imputation for Periodic Business Surveys\u201d. <em>Survey Methodology<\/em>, Vol 12, pp. 73-83.<\/p>\n<p>McGill, R., Tukey, J. W. and Larsen, W. A. (1978) \u201cVariations of box plots\u201d. <em>The American Statistician<\/em>, 32, pp. 12-16.<\/p>\n<p>Rousseeuw, P.J. and Croux, C. (1993) \u201cAlternatives to the Median Absolute Deviation\u201d, <em>Journal of the American Statistical Association<\/em>, 88, pp. 1273-1283.<\/p>\n<p>Hubert, M., and Vandervieren, E. (2008) \u201cAn Adjusted Boxplot for Skewed Distributions\u201d, <em>Computational Statistics &amp; Data Analysis<\/em>, 52, pp. 5186-5201<\/p>\n<p>&nbsp;<\/p>\n<p><strong>OTHER DOCUMENTATION<\/strong><\/p>\n<p><a href=\"https:\/\/github.com\/marcellodo\/univOutl\">https:\/\/github.com\/marcellodo\/univOutl<\/a><\/p>\n<p>&nbsp;<\/p>\n                                <\/div>\r\n                                                                                <\/div>\r\n                        \t\t\t\t\t<\/div>\r\n\t\t\t\t<\/div>\r\n\t\t\t\t\t<\/div>\r\n\t<\/div><\/div><\/section>","protected":false},"excerpt":{"rendered":"<p>The contents related to univOutl are shown in the following sections:<\/p>\n","protected":false},"author":19,"featured_media":0,"parent":2728,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_acf_changed":false,"footnotes":""},"class_list":["post-123984","page","type-page","status-publish","hentry"],"acf":[],"wpml_current_locale":"en_US","wpml_translations":[],"_links":{"self":[{"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/pages\/123984","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/users\/19"}],"replies":[{"embeddable":true,"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/comments?post=123984"}],"version-history":[{"count":4,"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/pages\/123984\/revisions"}],"predecessor-version":[{"id":124012,"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/pages\/123984\/revisions\/124012"}],"up":[{"embeddable":true,"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/pages\/2728"}],"wp:attachment":[{"href":"https:\/\/www.istat.it\/en\/wp-json\/wp\/v2\/media?parent=123984"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}