{"id":20227,"date":"2025-02-27T00:00:00","date_gmt":"2025-02-27T00:00:00","guid":{"rendered":"https:\/\/thierrymoudiki.github.io\/\/blog\/2025\/02\/27\/python\/r\/forecasting\/carbon-beta"},"modified":"2025-02-27T00:00:00","modified_gmt":"2025-02-27T00:00:00","slug":"presenting-online-probabilistic-estimation-of-carbon-beta-and-carbon-shapley-values-for-financial-and-climate-risk-at-institut-louis-bachelier","status":"publish","type":"post","link":"https:\/\/python-bloggers.com\/2025\/02\/presenting-online-probabilistic-estimation-of-carbon-beta-and-carbon-shapley-values-for-financial-and-climate-risk-at-institut-louis-bachelier\/","title":{"rendered":"Presenting &#8216;Online Probabilistic Estimation of Carbon Beta and Carbon Shapley Values for Financial and Climate Risk&#8217; at Institut Louis Bachelier"},"content":{"rendered":"<div style=\\\"border: 1px solid; background: none repeat scroll 0 0 #EDEDED; margin: 1px; font-size: 12px;\\\">\r\n<i>This article was first published on  <strong>\r\n<a href=\"https:\/\/thierrymoudiki.github.io\/\/blog\/2025\/02\/27\/python\/r\/forecasting\/carbon-beta\"> T. Moudiki's Webpage - Python <\/a><\/strong>, and kindly contributed to <a href=\/about\/>python-bloggers<\/a>.  (You can report issue about the content on this page <a href=\/contact-us\/>here<\/a>)\r\n<br\/>Want to share your content on python-bloggers?<a href=\/add-your-blog\/> click here<\/a>.<\/i>\r\n<\/div>\n<p>I\u2019ll be presenting my (preprint) paper, \u201cOnline Probabilistic Estimation of Carbon Beta and Carbon Shapley Values for Financial and Climate Risk,\u201d at the <a href=\"https:\/\/www.risks-forum.org\/\">18th FINANCIAL RISKS INTERNATIONAL FORUM<\/a>, hosted by <a href=\"https:\/\/www.institutlouisbachelier.org\/\">Institut Louis Bachelier<\/a> (link to the preprint at the end of the post).<\/p>\n<p>As climate change becomes a key financial risk factor, investors seek reliable ways to measure the exposure of stocks to climate transition risks. This paper introduces methods to estimate <strong>Carbon Beta<\/strong> and <strong>Carbon Shapley values<\/strong> dynamically and probabilistically.<\/p>\n<ul>\n<li><strong>Carbon Beta<\/strong> measures how stock returns react to a <em>Brown Minus Green (BMG)<\/em> portfolio, which holds long positions in carbon-intensive (brown) stocks and short positions in climate-friendly (green) stocks.<\/li>\n<li><strong>Carbon Shapley values<\/strong>, inspired by game theory, quantify the additive contribution of input factors to model predictions, helping explain stock return sensitivities.<\/li>\n<\/ul>\n<h2 id=\"context-from-capm-to-carbon-beta\">Context: From CAPM to Carbon Beta<\/h2>\n<p>The <strong>Capital Asset Pricing Model (CAPM)<\/strong> (Sharpe, 1964) introduced <strong>Beta<\/strong> as a measure of stock risk relative to the market. Over time, more sophisticated approaches\u2014machine learning (ML), neural networks, and game-theoretic Shapley values\u2014have emerged.<\/p>\n<p><strong>Carbon Beta<\/strong> extends this concept, capturing <strong>climate risk<\/strong> by analyzing how a stock\u2019s returns move with a BMG portfolio. A <strong>high Carbon Beta<\/strong> means a stock is highly exposed to the risks of transitioning to a low-carbon economy.<\/p>\n<h2 id=\"proposed-methodology\">Proposed Methodology<\/h2>\n<p><strong>Unlike traditional methods that assume a fixed, linear relationship between stock and market returns, the approach is described in the paper in **nonlinear**, **adaptive, nonparametric, and uncertainty-aware**.<\/strong><\/p>\n<p>My study introduces a <strong>machine learning (ML)-based, online estimation<\/strong> of Carbon Beta and Carbon Shapley values. Key innovations include:<\/p>\n<ol>\n<li><strong>No assumption of a \u201ctrue\u201d Carbon Beta<\/strong>\n<ul>\n<li>Uses numerical derivatives instead of a fixed linear model.<\/li>\n<\/ul>\n<\/li>\n<li><strong>Uncertainty quantification<\/strong>\n<ul>\n<li>Employs <strong>conformal prediction<\/strong> to provide confidence intervals around estimates.<\/li>\n<\/ul>\n<\/li>\n<li><strong>Model-agnostic Shapley values<\/strong>\n<ul>\n<li>Computes dynamic Shapley values to understand the influence of climate risk factors on stock returns.<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<p><strong>Link to the preprint:<\/strong><\/p>\n<p><a href=\"https:\/\/www.researchgate.net\/publication\/387577137_Online_probabilistic_estimation_of_carbon_beta_and_carbon_Shapley_values_for_financial_and_climate_risk\">https:\/\/www.researchgate.net\/publication\/387577137_Online_probabilistic_estimation_of_carbon_beta_and_carbon_Shapley_values_for_financial_and_climate_risk<\/a><\/p>\n<p><img data-recalc-dims=\"1\" decoding=\"async\" src=\"https:\/\/i0.wp.com\/thierrymoudiki.github.io\/images\/2025-02-27\/2025-02-27-image1.png?w=578&#038;ssl=1\" alt=\"image-title-here\" \/><\/p>\n\n<div style=\\\"border: 1px solid; background: none repeat scroll 0 0 #EDEDED; margin: 1px; font-size: 13px;\\\">\r\n<div style=\\\"text-align: center;\\\">To <strong>leave a comment<\/strong> for the author, please follow the link and comment on their blog: <strong><a href=\"https:\/\/thierrymoudiki.github.io\/\/blog\/2025\/02\/27\/python\/r\/forecasting\/carbon-beta\"> T. Moudiki's Webpage - Python <\/a><\/strong>.<\/div>\r\n<hr \/>\r\nWant to share your content on python-bloggers?<a href=\/add-your-blog\/ rel=\\\"nofollow\\\"> click here<\/a>.\r\n<\/div>","protected":false},"excerpt":{"rendered":"<div style = \"width: 60%; display: inline-block; float:left; \"> Presenting &#8216;Online Probabilistic Estimation of Carbon Beta and Carbon Shapley Values for Financial and Climate Risk&#8217; at Institut Louis Bachelier for the 18th FINANCIAL RISKS INTERNATIONAL FORUM<\/div>\n<div style = \"width: 40%; display: inline-block; float:right;\"><img id=\"excerpts_images\" src=' https:\/\/thierrymoudiki.github.io\/images\/2025-02-27\/2025-02-27-image1.png' width = \"200\"  style = \"padding: 10px;\" \/><\/div>\n<div style=\"clear: both;\"><\/div>\n","protected":false},"author":8,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[4],"tags":[],"class_list":["post-20227","post","type-post","status-publish","format-standard","hentry","category-data-science"],"aioseo_notices":[],"jetpack_featured_media_url":"","jetpack-related-posts":[],"jetpack_sharing_enabled":true,"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/posts\/20227","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/comments?post=20227"}],"version-history":[{"count":1,"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/posts\/20227\/revisions"}],"predecessor-version":[{"id":20228,"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/posts\/20227\/revisions\/20228"}],"wp:attachment":[{"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/media?parent=20227"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/categories?post=20227"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/python-bloggers.com\/wp-json\/wp\/v2\/tags?post=20227"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}