{"id":36537,"date":"2025-11-13T17:27:10","date_gmt":"2025-11-13T16:27:10","guid":{"rendered":"https:\/\/www.graviton.at\/letterswaplibrary\/mvp-building-a-llm-stock-forecasts-dataset-need-methodology-feedback\/"},"modified":"2025-11-13T17:27:10","modified_gmt":"2025-11-13T16:27:10","slug":"mvp-building-a-llm-stock-forecasts-dataset-need-methodology-feedback","status":"publish","type":"post","link":"https:\/\/www.graviton.at\/letterswaplibrary\/mvp-building-a-llm-stock-forecasts-dataset-need-methodology-feedback\/","title":{"rendered":"[MVP] Building A LLM Stock Forecasts Dataset (Need Methodology Feedback)"},"content":{"rendered":"<p><!-- SC_OFF --><\/p>\n<div class=\"md\">\n<p>Curious about LLM prediction performance, I built this tool to create an auditable, transparent record of LLM (GPT-Models) stock forecasts. I know LLMs aren&#8217;t designed for predictions, but I think a huge amount of data can give more insights into their capabilities.<\/p>\n<p>The core idea is methodology transparency:<\/p>\n<p>Baseline: The app ignores the price at the request time. It uses the Actual Closing Price (T_0) as the non-negotiable baseline for all subsequent sequential trend checks.<\/p>\n<p>Tracking: Accuracy is measured against both the Overall Trend (start to finish) and sequential Micro-Trends (step-by-step). Test a tracking run and critique the methodology:<\/p>\n<p><a href=\"https:\/\/glassballai.sumotrainer.com\/main\">https:\/\/glassballai.sumotrainer.com\/main<\/a><\/p>\n<p>Seeking Feedback: Does using the Closing Price as the sequential baseline feel robust for this type of analysis?<\/p>\n<p>Any other key input parameter (like specific news volume or market sentiment) I should be tracking?<\/p>\n<p>(Note: This is an MVP on a temporary URL and is not financial advice.)<\/p>\n<\/div>\n<p><!-- SC_ON -->   submitted by   <a href=\"https:\/\/www.reddit.com\/user\/aufgeblobt\"> \/u\/aufgeblobt <\/a> <br \/> <span><a href=\"https:\/\/www.reddit.com\/r\/datasets\/comments\/1ow59hp\/mvp_building_a_llm_stock_forecasts_dataset_need\/\">[link]<\/a><\/span>   <span><a href=\"https:\/\/www.reddit.com\/r\/datasets\/comments\/1ow59hp\/mvp_building_a_llm_stock_forecasts_dataset_need\/\">[comments]<\/a><\/span><\/p><div class='watch-action'><div class='watch-position align-right'><div class='action-like'><a class='lbg-style1 like-36537 jlk' href='javascript:void(0)' data-task='like' data-post_id='36537' data-nonce='65e0e39b87' rel='nofollow'><img class='wti-pixel' src='https:\/\/www.graviton.at\/letterswaplibrary\/wp-content\/plugins\/wti-like-post\/images\/pixel.gif' title='Like' \/><span class='lc-36537 lc'>0<\/span><\/a><\/div><\/div> <div class='status-36537 status align-right'><\/div><\/div><div class='wti-clear'><\/div>","protected":false},"excerpt":{"rendered":"<p>Curious about LLM prediction performance, I built this tool to create an auditable, transparent record of LLM&#8230;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[85],"tags":[],"class_list":["post-36537","post","type-post","status-publish","format-standard","hentry","category-datatards","wpcat-85-id"],"_links":{"self":[{"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/posts\/36537","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/comments?post=36537"}],"version-history":[{"count":0,"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/posts\/36537\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/media?parent=36537"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/categories?post=36537"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.graviton.at\/letterswaplibrary\/wp-json\/wp\/v2\/tags?post=36537"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}