{"id":2816,"date":"2026-04-16T08:06:23","date_gmt":"2026-04-16T08:06:23","guid":{"rendered":"https:\/\/blogs.mathworks.com\/finance\/?p=2816"},"modified":"2026-04-16T08:06:24","modified_gmt":"2026-04-16T08:06:24","slug":"whats-new-in-matlab-r2026a-for-economists","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/finance\/2026\/04\/16\/whats-new-in-matlab-r2026a-for-economists\/","title":{"rendered":"What&#8217;s New in MATLAB R2026a for Economists"},"content":{"rendered":"<p>R2026a covers a lot of ground for economists\u2014Bayesian state-space estimation, macro-scale forecasting, climate and physical risk mapping, symbolic dynamics, and AI-assisted model review, among others. <br \/>This post is not a comprehensive list. It is a short, practical tour through a few highlights that make economic modeling workflows easier to work with\u2014and easier to trust\u2014when models need to be reused rather than rewritten.  <\/p>\n<p>Links are included so you can go deeper into the areas that matter most to your work.<\/p>\n<p>If you would like to explore how these apply to your models, <strong>contact us to start a conversation.<\/strong><\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"is-layout-flex wp-block-buttons\">\n<div class=\"wp-block-button has-custom-font-size\" style=\"font-size:17px\"><a class=\"wp-block-button__link has-background wp-element-button\" href=\"mailto:centralbanks@mathworks.com\" style=\"background-color:#0076a8\" target=\"_blank\" rel=\"noreferrer noopener\">Contact Us<\/a><\/div>\n<\/div>\n<div style=\"height:50px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Build and Estimate Bayesian and State-Space Econometric Models<\/strong><\/h1>\n<p>R2026a includes a major update to Econometric Modeler, focusing on the full lifecycle of time\u2011series and state\u2011space analysis: from exploration and diagnostics, through simulation and forecasting.<\/p>\n<h2><strong>What\u2019s new<\/strong><\/h2>\n<ul>\n<li>A redesigned Econometric Modeler workflow with improved diagnostics, and support for deterministic and simulation-based forecasting<\/li>\n<li>Expanded non-linear state-space modeling, including new Sequential Monte Carlo proposal options such as auxiliary filtering and unscented Kalman filter\u2013based proposals<\/li>\n<li>New random effects panel data modeling with <span style=\"font-family: monospace\">fitrepanel<\/span><\/li>\n<\/ul>\n<p><strong>Example uses<\/strong><br \/>Macroeconomic foundations \u00b7 stochastic forecasts \u00b7 recession\/expansion regime analysis \u00b7 macro panel regression<\/p>\n<div style=\"height:20px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<figure class=\"wp-block-image size-large is-resized\"><img decoding=\"async\" loading=\"lazy\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Econ-Modeler-App-1024x920.png\" alt=\"Econometric Modeler app showing a time\u2011series and state\u2011space modeling workflow.\" class=\"wp-image-2818\" width=\"658\" height=\"590\"\/><figcaption class=\"wp-element-caption\">Econometric Modeler App screenshot<\/figcaption><\/figure>\n<div style=\"height:10px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<p><em>These updates are part of <strong><a rel=\"noreferrer noopener\" href=\"https:\/\/www.mathworks.com\/products\/econometrics.html\" target=\"_blank\">Econometrics Toolbox<\/a><\/strong><\/em><\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<hr class=\"wp-block-separator has-text-color has-alpha-channel-opacity has-background is-style-wide\" style=\"background-color:#b1b1b1;color:#b1b1b1\"\/>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Map Climate and Physical Risk to Financial Exposures<\/strong><\/h1>\n<p>Climate risk analysis often starts with spreadsheets and quickly turns into a geography problem.<\/p>\n<ul>\n<li>Where are the exposures located? <\/li>\n<li>How do hazards overlap with portfolios?<\/li>\n<li>How do you communicate those results clearly?<\/li>\n<\/ul>\n<h2><strong>What\u2019s new<\/strong><\/h2>\n<ul>\n<li>Display hazard rasters with <strong>geoimage<\/strong> and <strong>geopcolor<\/strong><\/li>\n<li>Build exposure heatmaps as pseudocolor rasters<\/li>\n<li>Add 3D OpenStreetMap buildings for geographical context<\/li>\n<li>Access expanded public hazard layers via WMS (Web Map Service)<\/li>\n<\/ul>\n<p><strong>Example uses<\/strong><br \/>Mortgage and commercial real estate (CRE) risk maps \u00b7 exposure hotspots \u00b7 climate risk reporting<\/p>\n<div style=\"height:20px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<figure class=\"wp-block-image size-full is-resized\"><img decoding=\"async\" loading=\"lazy\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Mapping-Toolbox.png\" alt=\"Map-based visualisation created with Mapping Toolbox, showing spatial financial exposures overlaid with climate or physical risk data.\" class=\"wp-image-2819\" width=\"668\" height=\"498\"\/><figcaption class=\"wp-element-caption\">CRE exposure around MathWorks Lakeside campus, with 3D buildings shaded by relative risk<\/figcaption><\/figure>\n<div style=\"height:10px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<p><em>These capabilities are provided by <strong><a rel=\"noreferrer noopener\" href=\"https:\/\/www.mathworks.com\/products\/mapping.html\" target=\"_blank\">Mapping Toolbox<\/a><\/strong><\/em><\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<hr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\/>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Integrate MATLAB into Agentic AI Workflows<\/strong><\/h1>\n<p>Econometric models are increasingly reviewed, tested, and discussed alongside AI tools, but that only works if the models themselves stay grounded, executable, and auditable.<\/p>\n<p>R2026a introduces a standardized way to integrate MATLAB into agentic AI workflows, so AI tools can interact with real models rather than static code snippets.<\/p>\n<h2><strong>What\u2019s new<\/strong><\/h2>\n<ul>\n<li>A standardized <strong>Model Context Protocol (MCP)<\/strong> interface for tools such as Claude, Gemini, Visual Studio Code, and GitHub Copilot<\/li>\n<li>Built-in tools for code execution, testing, and toolbox detection<\/li>\n<\/ul>\n<p><strong>Example uses<\/strong><br \/>Reviewing macroeconomic model code \u00b7 debugging state-space and forecasting workflows \u00b7 generating econometric analysis. <\/p>\n<div style=\"height:20px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<figure class=\"wp-block-video\"><video autoplay controls loop src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Claude-calls-MATLAB_.mp4\"><\/video><figcaption class=\"wp-element-caption\">Claude Calling MATLAB<\/figcaption><\/figure>\n<div style=\"height:10px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<p><em>This workflow is enabled by the <strong><a href=\"https:\/\/www.mathworks.com\/products\/matlab-mcp-core-server.html\" data-type=\"URL\" data-id=\"https:\/\/www.mathworks.com\/products\/matlab-mcp-core-server.html\" target=\"_blank\" rel=\"noreferrer noopener\">MATLAB MCP Core Server<\/a><\/strong><\/em><\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<hr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\/>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Analyze Economic Dynamics Symbolically<\/strong><\/h1>\n<p>Sometimes you don\u2019t want another simulation.<br \/>You want the equation.<\/p>\n<p>R2026a expands symbolic capabilities that support closed\u2011form analysis of dynamic systems\u2014useful for validating models, checking assumptions, and understanding outputs.<\/p>\n<h2><strong>What\u2019s new<\/strong><\/h2>\n<ul>\n<li>Solve linear recurrences relations and systems<\/li>\n<li>Support for selected non-linear and multiplicative forms<\/li>\n<li>Closed-form solutions with initial conditions, or use substitution to solve non-computable recurrence.<\/li>\n<\/ul>\n<p><strong>Example uses<\/strong><br \/>Dynamic stochastic general equilibrium (DSGE) or growth model analytic checks \u00b7 dynamic path formulas \u00b7 verifying iterative solvers<\/p>\n<div style=\"height:10px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<p><em>These features are available in <strong><a href=\"https:\/\/www.mathworks.com\/products\/symbolic.html\" data-type=\"URL\" data-id=\"https:\/\/www.mathworks.com\/products\/symbolic.html\" target=\"_blank\" rel=\"noreferrer noopener\">Symbolic Math Toolbox<\/a><\/strong><\/em><\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<hr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\/>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Prototype Time-Series Forecasts Without Writing Code<\/strong><\/h1>\n<p>Forecasting often starts with exploration.<br \/>What model works? How sensitive is it? Does a nonlinear approach help?<\/p>\n<p>R2026a continues to support low\u2011code forecasting workflows that let economists prototype quickly \u2014 and compare modern neural approaches with classical time\u2011series methods.<\/p>\n<h2><strong>What\u2019s new<\/strong><\/h2>\n<ul>\n<li>Build LSTM and other neural time-series models without code<\/li>\n<li>Compare neural forecasts with classical approaches such as ARIMA<\/li>\n<li>Training diagnostics, overfitting detection, and automatic code export<\/li>\n<\/ul>\n<p><strong>Example uses<\/strong><br \/>Nonlinear forecasting \u00b7 rapid prototyping<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<figure class=\"wp-block-image size-large is-resized\"><img decoding=\"async\" loading=\"lazy\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Time-series-modeler-1024x862.png\" alt=\"Screenshot of the Time Series Modeler app showing the data import and configuration step.\" class=\"wp-image-2826\" width=\"761\" height=\"640\"\/><figcaption class=\"wp-element-caption\">Importing and preparing time\u2011series data in the Time Series Modeler app<\/figcaption><\/figure>\n<div style=\"height:10px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<p><em>Time Series Modeler app is included in <strong><a rel=\"noreferrer noopener\" href=\"https:\/\/www.mathworks.com\/products\/deep-learning.html\" data-type=\"URL\" data-id=\"https:\/\/www.mathworks.com\/products\/deep-learning.html\" target=\"_blank\">Deep Learning Toolbox<\/a><\/strong><\/em><\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<hr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\/>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Enhanced Day-to-Day Research in MATLAB<\/strong><\/h1>\n<p>Some of the most valuable changes show up in the background \u2013 in data handling, interoperability, and performance. R2026a includes several updates that smooth the day\u2011to\u2011day work of economic research.<\/p>\n<p><strong>Data Access &amp; Interoperability<\/strong><\/p>\n<ul>\n<li>JSON import\/export to tables and timetables<\/li>\n<li>Unified Python environment management via the External Languages panel<\/li>\n<\/ul>\n<p><strong>Data Preparation &amp; Exploration<\/strong><\/p>\n<ul>\n<li>More flexible outlier detection in Live Scripts and the Data Cleaner app<\/li>\n<li>Improved missing\u2011data handling (mean, median, mode)<\/li>\n<li>Percentiles and quantiles for <span style=\"font-family: monospace\">datetime<\/span> and time\u2011series data<\/li>\n<\/ul>\n<p><strong>Modeling &amp; Numerical Analysis<\/strong><\/p>\n<ul>\n<li>Automatic differentiation for Jacobians in dynamic economic models<\/li>\n<li>Support for stiff and implicit dynamic systems (IDAS solver)<\/li>\n<\/ul>\n<p><strong>Visualization &amp; Communication<\/strong><\/p>\n<ul>\n<li>Raincloud plots for distributional analysis<\/li>\n<li>Interactive figures exportable as standalone web graphics<\/li>\n<li>Multilevel lists for clearer documentation in Live Scripts<\/li>\n<\/ul>\n<p><strong>Performance &amp; Productivity<\/strong><\/p>\n<ul>\n<li>Faster scatter plots<\/li>\n<li>Improved performance of core numerical functions (including log)<\/li>\n<li>Faster table joins and grouping for large datasets<\/li>\n<li>Faster MATLAB startup and everyday statistical workflows<\/li>\n<\/ul>\n<div style=\"height:10px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<p><em>These enhancements are part of <\/em><a rel=\"noreferrer noopener\" href=\"https:\/\/www.mathworks.com\/products\/matlab.html\" target=\"_blank\"><strong><em>Core MATLAB<\/em><\/strong><\/a><\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<hr class=\"wp-block-separator has-alpha-channel-opacity is-style-wide\"\/>\n<h1>Learn more<\/h1>\n<p>If you would like to go deeper into any of the workflows above, we would be happy to schedule an online meeting with you to discuss your specific modeling challenges.<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"is-layout-flex wp-block-buttons\">\n<div class=\"wp-block-button has-custom-font-size\" style=\"font-size:17px\"><a class=\"wp-block-button__link has-white-color has-text-color has-background wp-element-button\" href=\"mailto:centralbanks@mathworks.com\" style=\"background-color:#d78825\" target=\"_blank\" rel=\"noreferrer noopener\">Get in Touch<\/a><\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<div class=\"overview-image\"><img decoding=\"async\"  class=\"img-responsive\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Econ-Modeler-App-1024x920.png\" onError=\"this.style.display ='none';\" \/><\/div>\n<p>R2026a covers a lot of ground for economists\u2014Bayesian state-space estimation, macro-scale forecasting, climate and physical risk mapping, symbolic dynamics, and AI-assisted model review, among&#8230; <a class=\"read-more\" href=\"https:\/\/blogs.mathworks.com\/finance\/2026\/04\/16\/whats-new-in-matlab-r2026a-for-economists\/\">read more >><\/a><\/p>\n","protected":false},"author":233,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[31,16,37,4,34],"tags":[],"_links":{"self":[{"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/posts\/2816"}],"collection":[{"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/users\/233"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/comments?post=2816"}],"version-history":[{"count":20,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/posts\/2816\/revisions"}],"predecessor-version":[{"id":2850,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/posts\/2816\/revisions\/2850"}],"wp:attachment":[{"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/media?parent=2816"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/categories?post=2816"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/tags?post=2816"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}