{"id":2851,"date":"2026-05-05T08:09:58","date_gmt":"2026-05-05T08:09:58","guid":{"rendered":"https:\/\/blogs.mathworks.com\/finance\/?p=2851"},"modified":"2026-05-05T08:22:09","modified_gmt":"2026-05-05T08:22:09","slug":"prototype-time-series-forecasts-with-deep-learning-without-writing-code","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/finance\/2026\/05\/05\/prototype-time-series-forecasts-with-deep-learning-without-writing-code\/","title":{"rendered":"Prototype Time-Series Forecasts with Deep Learning\u2014Without Writing Code"},"content":{"rendered":"<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div>\n<table style=\"background-color: #e2f0ff;\">\n<tbody>\n<tr>\n<td style=\"width: 120px; padding: 3px; vertical-align: middle;\"><img decoding=\"async\" loading=\"lazy\" class=\"aligncenter wp-image-19243\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/05\/dong-yuchen-headshot.png\" alt=\"\" width=\"100\" height=\"100\"><\/td>\n<td style=\"vertical-align: middle; padding: 3px;\"><strong>Expert Contributor: <a href=\"https:\/\/www.linkedin.com\/in\/yuchen-dong-48061582\/\" target=\"_blank\" rel=\"noopener\">Dr. Yuchen Dong<\/a><\/strong><\/p>\n<h6><\/h6>\n<p><strong>Yuchen<\/strong> is a Senior Application Engineer at MathWorks focusing on customers in the financial services industry. His focus areas are financial instruments, portfolio optimization, and risk management. Before joining MathWorks, Yuchen worked as a derivative valuation analyst. He holds a Ph.D. in mathematical sciences and a master\u2019s degree in financial mathematics. <\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<div style=\"height:50px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<p>Forecasting workflows often begin with exploration:<\/p>\n<ul>\n<li>Does a nonlinear approach help?<\/li>\n<li>How sensitive are results to architecture or training choices?<\/li>\n<li>How do neural methods compare with classical benchmarks?<\/li>\n<\/ul>\n<p>The <strong>Time Series Modeler app<\/strong> in MATLAB is designed for this early phase. It lets you build, train, and compare time\u2011series forecasting models\u2014including deep learning approaches\u2014using a guided, visual workflow that complements code\u2011based analysis.<\/p>\n<p>This post walks through the main steps in the app, alongside the video, so you can follow or revisit each stage at your own pace.<\/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\"><a class=\"wp-block-button__link has-white-color has-text-color has-background wp-element-button\" href=\"https:\/\/www.mathworks.com\/videos\/no-code-forecasting-with-time-series-modeler-app-1777529946279.html\" style=\"background-color:#0076a8\" target=\"_blank\" rel=\"noreferrer noopener\">Watch the video<\/a><\/div>\n<\/div>\n<div style=\"height:50px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 1: Open the Time Series Modeler App<\/strong><\/h1>\n<p>To begin, open MATLAB and launch the <strong>Time Series Modeler<\/strong> app from the Apps tab. This opens an interactive environment for building time\u2011series forecasting models without writing code<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img decoding=\"async\" loading=\"lazy\" width=\"175\" height=\"202\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture1.png\" alt=\"image of the time series modeler app icon\" class=\"wp-image-2852\"\/><figcaption class=\"wp-element-caption\"><em>Time Series Modeler App icon<\/em><\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 2: Import Time-Series Data<\/strong><\/h1>\n<p>Import your dataset in <strong>timetable format<\/strong>. The app automatically detects the time variable and lists the remaining variables for selection.<\/p>\n<p>At this stage you can:<\/p>\n<ul>\n<li>Select predictors and response variables<\/li>\n<li>Specify the proportion of data used for validation<\/li>\n<\/ul>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img decoding=\"async\" loading=\"lazy\" src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture2-1014x1024.png\" alt=\"\" class=\"wp-image-2853\" width=\"658\" height=\"665\"\/><figcaption class=\"wp-element-caption\">Choose Brent as the response, select predictors, and define the validation split. <\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 3: Inspect and Explore the Data<\/strong><\/h1>\n<p>After import, the app displays diagnostic plots and summary information so you can explore the data before modeling.<\/p>\n<p>This step is useful for previewing the time-series data, checking observation structure, and getting an early sense of patterns before modeling. <\/p>\n<p>Spending a few moments here can prevent unnecessary retraining later.<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full is-resized\"><img decoding=\"async\" loading=\"lazy\" src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture3.png\" alt=\"\" class=\"wp-image-2854\" width=\"553\" height=\"468\"\/><figcaption class=\"wp-element-caption\">Data and time-series plots<\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 4: Select and Configure a Model<\/strong><\/h1>\n<p>To begin modeling, choose a predefined forecasting model from the <strong>Model<\/strong> section.<\/p>\n<p>For example, selecting <strong>LSTM (Small)<\/strong> opens a configuration window where you can:<\/p>\n<ul>\n<li>adjust key hyperparameters<\/li>\n<li>set training options<\/li>\n<li>choose solvers and learning rates<\/li>\n<li>view a diagram of the neural network architecture<\/li>\n<\/ul>\n<p>The app includes predefined deep learning models such as LSTM, GRU, MLP, and CNN netowkrs, and it also supports ARMA models for comparison when System Identification Toolbox is available. <\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img decoding=\"async\" loading=\"lazy\" src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture4-1024x555.png\" alt=\"\" class=\"wp-image-2855\" width=\"640\" height=\"347\"\/><figcaption class=\"wp-element-caption\">Model selection and configuration window<\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 5: Customize the Network (Optional)<\/strong><\/h1>\n<p>If your application requires a more specialized architecture, click Customize Network.<\/p>\n<p>This opens a simplified network editor based on Deep Network Designer, where you can:<\/p>\n<ul>\n<li>drag and drop layers from the library<\/li>\n<li>build a custom network structure<\/li>\n<li>visualize how data flows through the network<\/li>\n<\/ul>\n<p>This option is useful when experimenting beyond standard templates while still staying within a visual workflow.<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img decoding=\"async\" loading=\"lazy\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture5-688x1024.png\" alt=\"\" class=\"wp-image-2856\" width=\"469\" height=\"698\"\/><\/figure>\n<\/div>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large\"><img decoding=\"async\" loading=\"lazy\" width=\"1024\" height=\"543\" src=\"http:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture6-1024x543.png\" alt=\"\" class=\"wp-image-2857\"\/><figcaption class=\"wp-element-caption\">Deep Network Designer interface<\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 6: Train the model and Review Diagnostics<\/strong><\/h1>\n<p>Click <strong>Train<\/strong> to start the training process.<\/p>\n<p>As training progresses, the app displays metrics in real time and provides diagnostic tools to help assess model fit. For example, if signs of overfitting appear, the app surfaces recommendations and configuration options that can help improve generalization.<\/p>\n<p>This makes it easier to iterate on model choices without guessing.<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large\"><img decoding=\"async\" loading=\"lazy\" width=\"1024\" height=\"592\" src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture7-1024x592.png\" alt=\"\" class=\"wp-image-2859\"\/><figcaption class=\"wp-element-caption\">Training progress and diagnostics<\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 7: Generate Predictions and Evaluate Results<\/strong><\/h1>\n<p>After training completes, click <strong>Predict<\/strong> to generate predictions on the training or validation data. <\/p>\n<p>The app compares predicted values with the observed response and displays visual overlays to help you assess accuracy and behavior over time.<\/p>\n<p>This makes it straightforward to evaluate whether a model is capturing key dynamics before moving on.<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large\"><img decoding=\"async\" loading=\"lazy\" width=\"1024\" height=\"551\" src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture8-1024x551.png\" alt=\"\" class=\"wp-image-2860\"\/><figcaption class=\"wp-element-caption\">Predicting Brent over the validation window<\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Step 8: Export the Model and Generated Code<\/strong><\/h1>\n<p>When you\u2019re satisfied with the results, use <strong>Export<\/strong> to:<\/p>\n<ul>\n<li>Save the trained model as a struct in the MATLAB workspace<\/li>\n<li>Export results to the workspace and generate a live script for predicting on new data<\/li>\n<\/ul>\n<p>This allows you to reproduce results, extend the analysis programmatically, or integrate the model into larger forecasting pipelines.<\/p>\n<div style=\"height:30px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large\"><img decoding=\"async\" loading=\"lazy\" width=\"1024\" height=\"509\" src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture9_cropped-1-1024x509.png\" alt=\"\" class=\"wp-image-2895\"\/><figcaption class=\"wp-element-caption\">Export options and generated code preview<\/figcaption><\/figure>\n<\/div>\n<div style=\"height:60px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1><strong>Summary: A Guided Workflow for Forecasting \u2013 No Manual Coding Required<\/strong><\/h1>\n<p>Throughout this workflow, <strong>no manual coding is required<\/strong>. The Time Series Modeler app guides you step by step \u2014 from data import and exploration, through model configuration, training, diagnostics, and evaluation \u2014 using an interactive, visual interface.<\/p>\n<p>As you work, the app <strong>automatically generates a live script for prediction on new data when you export<\/strong>. This allows you to reproduce results, extend the analysis programmatically, or integrate the trained model into larger forecasting workflows when needed.<\/p>\n<p>The result is a workflow that supports rapid exploration and comparison early on, while still producing transparent, reusable artifacts for downstream use.<\/p>\n<p><em>The Time Series Modeler app is included in the Deep Learning Toolbox.<\/em><\/p>\n<\/p>\n<div style=\"height:50px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n<h1>Ready to try it yourself?<\/h1>\n<ul>\n<li>Explore related examples and documentation to go deeper into time-series forecasting with deep learning: <a rel=\"noreferrer noopener\" href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/timeseriesmodeler-app.html\" target=\"_blank\">Explore Time-Series Forecasting Workflows<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<div class=\"overview-image\"><img src=\"https:\/\/blogs.mathworks.com\/finance\/files\/2026\/04\/Picture3.png\" class=\"img-responsive attachment-post-thumbnail size-post-thumbnail wp-post-image\" alt=\"\" decoding=\"async\" loading=\"lazy\" \/><\/div>\n<p>Expert Contributor: Dr. Yuchen Dong<\/p>\n<p>Yuchen is a Senior Application Engineer at MathWorks focusing on customers in the financial services industry. His focus areas are financial instruments,&#8230; <a class=\"read-more\" href=\"https:\/\/blogs.mathworks.com\/finance\/2026\/05\/05\/prototype-time-series-forecasts-with-deep-learning-without-writing-code\/\">read more >><\/a><\/p>\n","protected":false},"author":237,"featured_media":2854,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[10,31,37,4,34],"tags":[],"_links":{"self":[{"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/posts\/2851"}],"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\/237"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/comments?post=2851"}],"version-history":[{"count":24,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/posts\/2851\/revisions"}],"predecessor-version":[{"id":2919,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/posts\/2851\/revisions\/2919"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/media\/2854"}],"wp:attachment":[{"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/media?parent=2851"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/categories?post=2851"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/finance\/wp-json\/wp\/v2\/tags?post=2851"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}