{"id":913,"date":"2018-04-22T17:21:18","date_gmt":"2018-04-22T17:21:18","guid":{"rendered":"http:\/\/www.skampakis.com\/?p=913"},"modified":"2023-05-10T08:36:03","modified_gmt":"2023-05-10T08:36:03","slug":"performance-measures-standard-deviation-mad-and-quantiles","status":"publish","type":"post","link":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/","title":{"rendered":"Performance measures: standard deviation, MAD, and quantiles"},"content":{"rendered":"

xSomething else which I have found helps a lot is the use of error histograms and the standard deviation or quantiles when reporting errors. So, for example,<\/p>\n

<GRAPH><\/p>\n

When a graph is relatively normal, the use of standard deviation makes sense, otherwise you might want to go for quantiles, or the median absolute deviation (MAD).<\/p>\n","protected":false},"excerpt":{"rendered":"

xSomething else which I have found helps a lot is the use of error histograms and the standard deviation or quantiles when reporting errors. So, for example, <GRAPH> When a graph is relatively normal, the use of standard deviation makes sense, otherwise you might want to go for quantiles, or the median absolute deviation (MAD).<\/p>\n","protected":false},"author":7,"featured_media":3622,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"_uf_show_specific_survey":0,"_uf_disable_surveys":false,"neve_meta_sidebar":"","neve_meta_container":"","neve_meta_enable_content_width":"","neve_meta_content_width":0,"neve_meta_title_alignment":"","neve_meta_author_avatar":"","neve_post_elements_order":"","neve_meta_disable_header":"","neve_meta_disable_footer":"","neve_meta_disable_title":"","_ti_tpc_template_sync":false,"_ti_tpc_template_id":"","_jetpack_memberships_contains_paid_content":false,"footnotes":"","_wpscppro_custom_social_share_image":0,"_facebook_share_type":"","_twitter_share_type":"","_linkedin_share_type":"","_pinterest_share_type":"","_linkedin_share_type_page":"","_selected_social_profile":[]},"categories":[17,16],"tags":[],"yoast_head":"\nPerformance measures: standard deviation, MAD, and quantiles - The Data Scientist<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Performance measures: standard deviation, MAD, and quantiles - The Data Scientist\" \/>\n<meta property=\"og:description\" content=\"xSomething else which I have found helps a lot is the use of error histograms and the standard deviation or quantiles when reporting errors. So, for example, <GRAPH> When a graph is relatively normal, the use of standard deviation makes sense, otherwise you might want to go for quantiles, or the median absolute deviation (MAD).\" \/>\n<meta property=\"og:url\" content=\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/\" \/>\n<meta property=\"og:site_name\" content=\"The Data Scientist\" \/>\n<meta property=\"article:published_time\" content=\"2018-04-22T17:21:18+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-05-10T08:36:03+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1280\" \/>\n\t<meta property=\"og:image:height\" content=\"854\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Dr. Stylianos (Stelios) Kampakis\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@s_kampakis\" \/>\n<meta name=\"twitter:site\" content=\"@s_kampakis\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Dr. Stylianos (Stelios) Kampakis\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"1 minute\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/\"},\"author\":{\"name\":\"Dr. Stylianos (Stelios) Kampakis\",\"@id\":\"https:\/\/thedatascientist.com\/#\/schema\/person\/787881a641bab92a8a46fcb198f9f6ba\"},\"headline\":\"Performance measures: standard deviation, MAD, and quantiles\",\"datePublished\":\"2018-04-22T17:21:18+00:00\",\"dateModified\":\"2023-05-10T08:36:03+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/\"},\"wordCount\":64,\"publisher\":{\"@id\":\"https:\/\/thedatascientist.com\/#organization\"},\"image\":{\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg\",\"articleSection\":[\"Machine learning\",\"Technical Posts\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/\",\"url\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/\",\"name\":\"Performance measures: standard deviation, MAD, and quantiles - The Data Scientist\",\"isPartOf\":{\"@id\":\"https:\/\/thedatascientist.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg\",\"datePublished\":\"2018-04-22T17:21:18+00:00\",\"dateModified\":\"2023-05-10T08:36:03+00:00\",\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage\",\"url\":\"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg\",\"contentUrl\":\"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg\",\"width\":1280,\"height\":854},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/thedatascientist.com\/#website\",\"url\":\"https:\/\/thedatascientist.com\/\",\"name\":\"The Data Scientist\",\"description\":\"data science courses, data science services, tokenomics consulting\",\"publisher\":{\"@id\":\"https:\/\/thedatascientist.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/thedatascientist.com\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/thedatascientist.com\/#organization\",\"name\":\"The Data Scientist\",\"url\":\"https:\/\/thedatascientist.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/thedatascientist.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/thedatascientist.com\/wp-content\/uploads\/2023\/05\/logo-TDS-white-text.webp\",\"contentUrl\":\"https:\/\/thedatascientist.com\/wp-content\/uploads\/2023\/05\/logo-TDS-white-text.webp\",\"width\":797,\"height\":249,\"caption\":\"The Data Scientist\"},\"image\":{\"@id\":\"https:\/\/thedatascientist.com\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/s_kampakis\",\"https:\/\/www.linkedin.com\/in\/dr-stylianos-kampakis\/\",\"https:\/\/www.youtube.com\/channel\/UClyTBR1b0ZEcZgWEhFf2POg\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/thedatascientist.com\/#\/schema\/person\/787881a641bab92a8a46fcb198f9f6ba\",\"name\":\"Dr. Stylianos (Stelios) Kampakis\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/thedatascientist.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/37d2748daf20ea5aa764cf8726fa34d1?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/37d2748daf20ea5aa764cf8726fa34d1?s=96&d=mm&r=g\",\"caption\":\"Dr. Stylianos (Stelios) Kampakis\"}}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Performance measures: standard deviation, MAD, and quantiles - The Data Scientist","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/","og_locale":"en_US","og_type":"article","og_title":"Performance measures: standard deviation, MAD, and quantiles - The Data Scientist","og_description":"xSomething else which I have found helps a lot is the use of error histograms and the standard deviation or quantiles when reporting errors. So, for example, <GRAPH> When a graph is relatively normal, the use of standard deviation makes sense, otherwise you might want to go for quantiles, or the median absolute deviation (MAD).","og_url":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/","og_site_name":"The Data Scientist","article_published_time":"2018-04-22T17:21:18+00:00","article_modified_time":"2023-05-10T08:36:03+00:00","og_image":[{"width":1280,"height":854,"url":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg","type":"image\/jpeg"}],"author":"Dr. Stylianos (Stelios) Kampakis","twitter_card":"summary_large_image","twitter_creator":"@s_kampakis","twitter_site":"@s_kampakis","twitter_misc":{"Written by":"Dr. Stylianos (Stelios) Kampakis","Est. reading time":"1 minute"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#article","isPartOf":{"@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/"},"author":{"name":"Dr. Stylianos (Stelios) Kampakis","@id":"https:\/\/thedatascientist.com\/#\/schema\/person\/787881a641bab92a8a46fcb198f9f6ba"},"headline":"Performance measures: standard deviation, MAD, and quantiles","datePublished":"2018-04-22T17:21:18+00:00","dateModified":"2023-05-10T08:36:03+00:00","mainEntityOfPage":{"@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/"},"wordCount":64,"publisher":{"@id":"https:\/\/thedatascientist.com\/#organization"},"image":{"@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage"},"thumbnailUrl":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg","articleSection":["Machine learning","Technical Posts"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/","url":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/","name":"Performance measures: standard deviation, MAD, and quantiles - The Data Scientist","isPartOf":{"@id":"https:\/\/thedatascientist.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage"},"image":{"@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage"},"thumbnailUrl":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg","datePublished":"2018-04-22T17:21:18+00:00","dateModified":"2023-05-10T08:36:03+00:00","inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/thedatascientist.com\/performance-measures-standard-deviation-mad-and-quantiles\/#primaryimage","url":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg","contentUrl":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg","width":1280,"height":854},{"@type":"WebSite","@id":"https:\/\/thedatascientist.com\/#website","url":"https:\/\/thedatascientist.com\/","name":"The Data Scientist","description":"data science courses, data science services, tokenomics consulting","publisher":{"@id":"https:\/\/thedatascientist.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/thedatascientist.com\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/thedatascientist.com\/#organization","name":"The Data Scientist","url":"https:\/\/thedatascientist.com\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/thedatascientist.com\/#\/schema\/logo\/image\/","url":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2023\/05\/logo-TDS-white-text.webp","contentUrl":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2023\/05\/logo-TDS-white-text.webp","width":797,"height":249,"caption":"The Data Scientist"},"image":{"@id":"https:\/\/thedatascientist.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/s_kampakis","https:\/\/www.linkedin.com\/in\/dr-stylianos-kampakis\/","https:\/\/www.youtube.com\/channel\/UClyTBR1b0ZEcZgWEhFf2POg"]},{"@type":"Person","@id":"https:\/\/thedatascientist.com\/#\/schema\/person\/787881a641bab92a8a46fcb198f9f6ba","name":"Dr. Stylianos (Stelios) Kampakis","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/thedatascientist.com\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/37d2748daf20ea5aa764cf8726fa34d1?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/37d2748daf20ea5aa764cf8726fa34d1?s=96&d=mm&r=g","caption":"Dr. Stylianos (Stelios) Kampakis"}}]}},"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"https:\/\/thedatascientist.com\/wp-content\/uploads\/2018\/04\/pexels-nothing-ahead-7059613.jpg","_links":{"self":[{"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/posts\/913"}],"collection":[{"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/comments?post=913"}],"version-history":[{"count":1,"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/posts\/913\/revisions"}],"predecessor-version":[{"id":3500,"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/posts\/913\/revisions\/3500"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/media\/3622"}],"wp:attachment":[{"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/media?parent=913"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/categories?post=913"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/thedatascientist.com\/wp-json\/wp\/v2\/tags?post=913"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}