{"id":67535,"date":"2024-02-09T11:57:06","date_gmt":"2024-02-09T19:57:06","guid":{"rendered":"https:\/\/drduru.com\/onetwentytwo\/?p=67535"},"modified":"2024-02-10T20:41:53","modified_gmt":"2024-02-11T04:41:53","slug":"how-to-use-chatgpt-to-test-theories-on-stock-prices","status":"publish","type":"post","link":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/","title":{"rendered":"How to Use ChatGPT to Test Theories on Stock Prices"},"content":{"rendered":"\n<p>I have a good friend that comes up with interesting theories about relationships in the stock market. Recently, he claimed that &#8220;more often than not&#8221;, on a day where the big cap stocks gain and the iShares Russell 2000 ETF (IWM) does not gain, IWM will experience a daily gain on the next trading day. He lamented that he could no longer investigate these theories, and he asked whether I could do it.<\/p>\n\n\n\n<p>I decided to test his theory with the help of &#8220;Data Analyst&#8221; in ChatGPT. My observant and detail-oriented friend was CORRECT! However, his observation comes with a HUGE caveat. While it is true IWM gained &#8220;more often than not&#8221; in this scenario, the gains do not meet the threshold of statistical significance. In other words, we cannot count on these IWM gains in the future. If a trader captures such gains, it could be the result of sheer luck.<\/p>\n\n\n\n<p>Despite the statistical disappointment, the test of this stock market theory provides a great example of the potential of ChatGPT for navigating the complexities of price dynamics in the stock market and to test theories on stock prices.<\/p>\n\n\n\n<br><script async=\"\" src=\"https:\/\/pagead2.googlesyndication.com\/pagead\/js\/adsbygoogle.js\"><\/script>\n<ins class=\"adsbygoogle\" style=\"display:block; text-align:center;\" data-ad-layout=\"in-article\" data-ad-format=\"fluid\" data-ad-client=\"ca-pub-1196033822186410\" data-ad-slot=\"3478942445\"><\/ins>\n<script>\n     (adsbygoogle = window.adsbygoogle || []).push({});\n<\/script><br>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>ChatGPT Prompt Engineering<\/strong><\/h2>\n\n\n\n<p>I used the Invesco QQQ Trust (QQQ) as a proxy for &#8220;big cap stocks&#8221; given this tech-laden ETF is <a href=\"https:\/\/swingtradebot.com\/equities\/QQQ\/etf-components\">dominated by the biggest tech stocks<\/a>. I downloaded the daily price data from Yahoo Finance <a href=\"https:\/\/finance.yahoo.com\/quote\/QQQ\/history?p=QQQ\">for QQQ<\/a> and <a href=\"https:\/\/finance.yahoo.com\/quote\/IWM\/history?p=IWM\">for IWM<\/a> for the last 5 years.<\/p>\n\n\n\n<p>My ChatGPT subscription gives me access to the Data Analyst feature. I uploaded manually merged csv files from Yahoo Finance and truncated the data down to closing prices. Here is my prompt for ChatGPT:<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p>&#8220;I have a theory that I want to test. On days where QQQ experiences a daily gain and IWM does not experience a daily gain, I believe, more often than not, IWM experiences a daily gain on the next trading day. Call this next trading day Scenario 1. Analyze this theory using the attached time series data of closing prices for IWM and QQQ. Calculate the percentage of days IWM experiences a gain excluding the trading days in Scenario 1. Call this Scenario X1. Include three different tests of statistical significance for Scenario 1 using Scenario X1 as the baseline. Also calculate the percentage of days IWM experiences a gain on the second or third trading day after the day QQQ gains and IWM does not. Call these Scenario 2 and Scenario 3 respectively.&#8221;<\/p>\n<\/blockquote>\n\n\n\n<p>I did not include all my analytics requests. I discovered that ChatGPT ran into all sorts of strange errors when presented with too many requests at the same. ChatGPT seems to perform best with an iterative conversation and with data ready to analyze. As a result, I started with the basic request above. I even had to experiment with different ways of stating and ordering my instructions. Needless to say I was surprised to discover the ordering made a difference in the accuracy of ChatGPT&#8217;s results! I assume ChatGPT interprets instructions best when they follow a logical order. The prompt engineering apparently should use a modular approach to problem-solving.<\/p>\n\n\n\n<p>Importantly, I did my own manual analysis to validate ChatGPT&#8217;s results (<a href=\"https:\/\/docs.google.com\/spreadsheets\/d\/1XLVilNf_UVze8zhw9c9NGKOCKR5RQSeMaEcPc_sY-Ho\/edit?usp=sharing\">data and analysis here<\/a>). This validation helped me discover that ChatGPT can work better when key concepts carry a label (think about creating labels in a machine-learning problem). ChatGPT calculate the correct percentage of gains for the baseline only after I gave it a label (&#8220;Scenario X1&#8221;).<\/p>\n\n\n\n<p>I also first gave ChatGPT the data in two separate files, one for QQQ and one for IWM, but the computing sweat required to merge the files seemed to degrade the quality of ChatGPT&#8217;s analysis. The Data Analyst at times even had trouble processing the files.<\/p>\n\n\n\n<p>I asked ChatGPT these follow-up questions:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li>Provide a bar chart showing the IWM percentage gains for Scenario 1, Scenario X1, Scenario 2, and Scenario 3.<\/li>\n\n\n\n<li>Redo the entire analysis. This time exclude trading data before 2023.<\/li>\n<\/ol>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>The Results Summary<\/strong><\/h2>\n\n\n\n<p>While IWM indeed gains &#8220;more often than not&#8221; in my friend&#8217;s scenario, the percentage of days with a gain are not different enough from the percentage of other days with a gain to generate statistical significance. In other words, the odds that the observed gains are the result of chance are too high to accept my friend&#8217;s observation as something to expect in the future.<\/p>\n\n\n\n<p>In the chart below, note the proximity of Scenario 1 (53.5%) and Scenario X1 (51.6%). Intuitively, I can see that the two percentage point difference is not significant and not enough of an edge for a meaningful trade. In other words, I might as well just buy and sell IWM every day (excluding the impact of the magnitude of gains and losses of course).<\/p>\n\n\n\n<p><\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><a href=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\"><img loading=\"lazy\" decoding=\"async\" width=\"806\" height=\"495\" src=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\" alt=\"Percentage of days IWM experiences a gain based on the combined performance of QQQ and IWM (February 12, 2019 to February 8, 2024)\" class=\"wp-image-67545\" srcset=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png 806w, https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data-768x472.png 768w\" sizes=\"auto, (max-width: 806px) 100vw, 806px\" \/><\/a><figcaption class=\"wp-element-caption\"><strong><em>Percentage of days IWM experiences a gain based on the combined performance of QQQ and IWM (February 12, 2019 to February 8, 2024)<\/em><\/strong><\/figcaption><\/figure>\n<\/div>\n\n\n<p>The tests of statistical significance confirm what my eyes tell me. From ChatGPT:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Z-test for Proportions<\/strong>: The p-value is 0.633, indicating that there is no statistically significant difference in the proportion of days IWM experiences a gain on the next trading day after Scenario 1 compared to other days (Scenario X1).<\/li>\n\n\n\n<li><strong>Chi-square Test for Independence<\/strong>: The p-value is 0.691, suggesting that there is no statistically significant association between the scenarios (Scenario 1 vs. Scenario X1) and the likelihood of IWM experiencing a gain.<\/li>\n\n\n\n<li><strong>T-test for the Means of Daily Percentage Changes<\/strong>: The p-value is 0.454, indicating that there is no statistically significant difference in the means of daily percentage changes for IWM between the next day of Scenario 1 and other days (Scenario X1).<\/li>\n<\/ol>\n\n\n\n<p>The threshold for statistical significance is conventionally a p-value below 0.05.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>The Results for 2023-2024<\/strong><\/h2>\n\n\n\n<p>Since my friend recently made this observation of the behavior of IWM versus QQQ, perhaps limiting the data to just the last year and this year would reveal more significance. While the percentage of days with a gain increased to 57.1% versus the baseline 50.0%, the observation still failed the tests of statistical significance. The observation was very real, but the effect it is not actionable.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"787\" height=\"501\" src=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-2023-2024-trading-data.png\" alt=\"Percentage of days IWM experiences a gain based on the combined performance of QQQ and IWM (January 3, 2023 to February 8, 2024)\" class=\"wp-image-67546\" srcset=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-2023-2024-trading-data.png 787w, https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-2023-2024-trading-data-768x489.png 768w\" sizes=\"auto, (max-width: 787px) 100vw, 787px\" \/><figcaption class=\"wp-element-caption\"><strong><em>Percentage of days IWM experiences a gain based on the combined performance of QQQ and IWM (January 3, 2023 to February 8, 2024)<\/em><\/strong><\/figcaption><\/figure>\n<\/div>\n\n\n<p>Despite the larger difference between Scenario 1 and Scenario X1, the statistical tests all definitively failed. The bar for significance is in a sense higher because of the smaller sample size. From ChatGPT:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Z-test for Proportions<\/strong>: The p-value is 0.364, suggesting that there is no statistically significant difference in the proportion of days IWM experiences a gain on the next trading day after Scenario 1 compared to other days (Scenario X1), within the 2023 data.<\/li>\n\n\n\n<li><strong>Chi-square Test for Independence<\/strong>: The p-value is 0.453, indicating that there is no statistically significant association between the scenarios (Scenario 1 vs. Scenario X1) and the likelihood of IWM experiencing a gain, for data from 2023 onwards.<\/li>\n\n\n\n<li><strong>T-test for the Means of Daily Percentage Changes<\/strong>: The p-value is 0.348, showing that there is no statistically significant difference in the means of daily percentage changes for IWM between the next day of Scenario 1 and other days (Scenario X1), within the 2023 data.<\/li>\n<\/ol>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-full\"><a href=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-intraday-iShares-Russell-2000-ETF.png\"><img loading=\"lazy\" decoding=\"async\" width=\"1480\" height=\"836\" src=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-intraday-iShares-Russell-2000-ETF.png\" alt=\"The iShares Russell 2000 ETF (IWM) has struggled to break out of a trading range and away from its bear market line. With no discernible trend, trading patterns beyond the upper and lower bounds of the range should be hard to find.\" class=\"wp-image-67553\" title=\"iShares Russell 2000 ETF (IWM)\" srcset=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-intraday-iShares-Russell-2000-ETF.png 1480w, https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-intraday-iShares-Russell-2000-ETF-768x434.png 768w, https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-intraday-iShares-Russell-2000-ETF-1320x746.png 1320w\" sizes=\"auto, (max-width: 1480px) 100vw, 1480px\" \/><\/a><figcaption class=\"wp-element-caption\"><strong><em>The iShares Russell 2000 ETF (IWM) has struggled to break out of a trading range and away from its bear market line. With no discernible trend, trading patterns beyond the upper and lower bounds of the range should be hard to find.<\/em><\/strong><\/figcaption><\/figure>\n<\/div>\n\n\n<h2 class=\"wp-block-heading\"><strong>Conclusion<\/strong><\/h2>\n\n\n\n<p>The Data Analyst in ChatGPT looks like a convenient way to quickly test theories on stock prices. However, my first attempt took a lot of prompt engineering before I got the desired behaviors and accuracy. I learned that ChatGPT appears to perform best with very detailed and tightly constrained requests. An iterative conversation supports ChatGPT&#8217;s need for detail and constraint. I expect to use these learnings to conduct faster tests in the future.<\/p>\n\n\n\n<p>Analysts often skip tests of statistical significance when mining for relationships in the stock market. ChatGPT&#8217;s Data Analyst makes these tests easy to access, execute and interpret. Thus, there is little excuse anymore for skipping this important step. Now, almost any stock market data mining exercise can include the context of significance and quantified relevance.<\/p>\n\n\n\n<p>Be careful out there!<\/p>\n\n\n\n<p>Full disclosure: long IWM calendar put spread<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><strong>Epilogue: ChatGPT&#8217;s Response<\/strong><\/h2>\n\n\n\n<p>Here is how ChatGPT responded to my first prompt&#8230;.<\/p>\n\n\n\n<p>To analyze your theory, I&#8217;ll proceed with the following steps:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Load and Inspect the Data<\/strong>: First, I&#8217;ll load the attached CSV file to understand its structure and the data it contains.<\/li>\n\n\n\n<li><strong>Calculate Daily Gains<\/strong>: Determine the daily percentage change for both QQQ and IWM.<\/li>\n\n\n\n<li><strong>Identify Scenario 1 Days<\/strong>: Days where QQQ experiences a daily gain and IWM does not.<\/li>\n\n\n\n<li><strong>Analyze Scenario 1<\/strong>: Calculate the percentage of days IWM experiences a gain on the next trading day after Scenario 1.<\/li>\n\n\n\n<li><strong>Calculate Scenario X1<\/strong>: Calculate the percentage of days IWM experiences a gain excluding the trading days in Scenario 1.<\/li>\n\n\n\n<li><strong>Statistical Significance Tests<\/strong>: Perform three different statistical tests to compare the results of Scenario 1 with Scenario X1.<\/li>\n\n\n\n<li><strong>Analyze Scenarios 2 and 3<\/strong>: Calculate the percentage of days IWM experiences a gain on the second and third trading day after the day QQQ gains and IWM does not.<\/li>\n<\/ol>\n\n\n\n<p>Let&#8217;s start with the first step: loading and inspecting the data.<\/p>\n\n\n\n<p>The data consists of three columns: <code>Date<\/code>, <code>QQQ<\/code> (closing prices for QQQ), and <code>IWM<\/code> (closing prices for IWM). With this understanding, I&#8217;ll proceed to calculate the daily percentage change for both QQQ and IWM to identify days that meet the criteria for Scenario 1 (QQQ gains, IWM does not gain), and then analyze the subsequent scenarios and perform statistical tests as outlined. \u200b\u200b<\/p>\n\n\n\n<p>Based on the analysis of the provided data, here are the findings:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Scenario 1<\/strong>: On the next trading day after QQQ experiences a daily gain and IWM does not, IWM experiences a daily gain <strong>53.48%<\/strong> of the time.<\/li>\n\n\n\n<li><strong>Scenario X1<\/strong>: Excluding the trading days in Scenario 1, IWM experiences a daily gain <strong>51.59%<\/strong> of the time.<\/li>\n<\/ul>\n\n\n\n<p>This indicates that following the criteria of Scenario 1, there is a slightly higher likelihood of IWM experiencing a gain on the next trading day compared to other days.<\/p>\n\n\n\n<p>Now, for the statistical significance tests comparing Scenario 1 with Scenario X1, I&#8217;ll perform three different tests:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Z-test<\/strong> for proportions,<\/li>\n\n\n\n<li><strong>Chi-square test<\/strong> for independence, and<\/li>\n\n\n\n<li><strong>T-test<\/strong> for the means of daily percentage changes.<\/li>\n<\/ol>\n\n\n\n<p>Additionally, the percentage of days IWM experiences a gain on the second and third trading day after the day QQQ gains and IWM does not (Scenarios 2 and 3) are:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Scenario 2<\/strong>: 50.80%<\/li>\n\n\n\n<li><strong>Scenario 3<\/strong>: 47.06%<\/li>\n<\/ul>\n\n\n\n<p>Next, I&#8217;ll conduct the statistical significance tests for Scenario 1 using Scenario X1 as the baseline. \u200b\u200b<\/p>\n\n\n\n<p>The results of the three statistical significance tests comparing Scenario 1 with Scenario X1 are as follows:<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>Z-test for Proportions<\/strong>: The p-value is 0.633, indicating that there is no statistically significant difference in the proportion of days IWM experiences a gain on the next trading day after Scenario 1 compared to other days (Scenario X1).<\/li>\n\n\n\n<li><strong>Chi-square Test for Independence<\/strong>: The p-value is 0.691, suggesting that there is no statistically significant association between the scenarios (Scenario 1 vs. Scenario X1) and the likelihood of IWM experiencing a gain.<\/li>\n\n\n\n<li><strong>T-test for the Means of Daily Percentage Changes<\/strong>: The p-value is 0.454, indicating that there is no statistically significant difference in the means of daily percentage changes for IWM between the next day of Scenario 1 and other days (Scenario X1).<\/li>\n<\/ol>\n\n\n\n<p>These statistical tests suggest that the difference observed in the percentage of days IWM experiences a gain following Scenario 1 compared to Scenario X1 does not have statistical significance. Thus, while there is a slight difference in the observed percentages, this difference is not enough to reject the null hypothesis of no difference between the scenarios under the conditions tested.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>I have a good friend that comes up with interesting theories about relationships in the stock market. Recently, he claimed that &#8220;more often than not&#8221;, on a day where the big cap stocks gain and the iShares Russell 2000 ETF (IWM) does not gain, IWM will experience a daily gain on the next trading day. &#8230; <a title=\"How to Use ChatGPT to Test Theories on Stock Prices\" class=\"read-more\" href=\"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/\">Read more<\/a><\/p>\n","protected":false},"author":1,"featured_media":67545,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[3252,135,62,1210,982],"tags":[3188,3422,3189,3187,2218,909,3423,535,3424],"class_list":["post-67535","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-artificial-intelligence","category-stock-market-trading","category-technical-analysis","category-trading-model","category-trading-range","tag-chatgpt","tag-data-analyst","tag-generative-ai","tag-invesco-qqq-trust","tag-ishares-russell-2000-etf","tag-iwm","tag-prompt-engineering","tag-qqq","tag-statistical-significance"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>How to Use ChatGPT to Test Theories on Stock Prices<\/title>\n<meta name=\"description\" content=\"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"How to Use ChatGPT to Test Theories on Stock Prices\" \/>\n<meta property=\"og:description\" content=\"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/\" \/>\n<meta property=\"og:site_name\" content=\"ONE-TWENTY TWO: Trading Financial Markets\" \/>\n<meta property=\"article:published_time\" content=\"2024-02-09T19:57:06+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-02-11T04:41:53+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\" \/>\n\t<meta property=\"og:image:width\" content=\"806\" \/>\n\t<meta property=\"og:image:height\" content=\"495\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Dr. Duru\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:title\" content=\"How to Use ChatGPT to Test Theories on Stock Prices\" \/>\n<meta name=\"twitter:description\" content=\"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.\" \/>\n<meta name=\"twitter:image\" content=\"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\" \/>\n<meta name=\"twitter:creator\" content=\"@drduru\" \/>\n<meta name=\"twitter:site\" content=\"@drduru\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Dr. Duru\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"9 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/\"},\"author\":{\"name\":\"Dr. Duru\",\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/#\\\/schema\\\/person\\\/d9569fde6c701b021a8d958f775be9a0\"},\"headline\":\"How to Use ChatGPT to Test Theories on Stock Prices\",\"datePublished\":\"2024-02-09T19:57:06+00:00\",\"dateModified\":\"2024-02-11T04:41:53+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/\"},\"wordCount\":1915,\"commentCount\":8,\"publisher\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/#\\\/schema\\\/person\\\/d9569fde6c701b021a8d958f775be9a0\"},\"image\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2024\\\/02\\\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\",\"keywords\":[\"ChatGPT\",\"data analyst\",\"generative AI\",\"Invesco QQQ Trust\",\"iShares Russell 2000 ETF\",\"IWM\",\"prompt engineering\",\"QQQ\",\"statistical significance\"],\"articleSection\":[\"Artificial Intelligence\",\"Stock Market Trading\",\"Technical Analysis\",\"Trading Model\",\"Trading Range\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/\",\"url\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/\",\"name\":\"How to Use ChatGPT to Test Theories on Stock Prices\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2024\\\/02\\\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\",\"datePublished\":\"2024-02-09T19:57:06+00:00\",\"dateModified\":\"2024-02-11T04:41:53+00:00\",\"description\":\"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#primaryimage\",\"url\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2024\\\/02\\\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\",\"contentUrl\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2024\\\/02\\\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png\",\"width\":806,\"height\":495,\"caption\":\"Percentage of days IWM experiences a gain based on the combined performance of QQQ and IWM (February 12, 2019 to February 8, 2024)\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/2024\\\/02\\\/09\\\/how-to-use-chatgpt-to-test-theories-on-stock-prices\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"How to Use ChatGPT to Test Theories on Stock Prices\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/#website\",\"url\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/\",\"name\":\"ONE-TWENTY TWO: Trading Financial Markets\",\"description\":\"Exploring the poetry in financial markets for students of money and economy since the year 2000.\",\"publisher\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/#\\\/schema\\\/person\\\/d9569fde6c701b021a8d958f775be9a0\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/#\\\/schema\\\/person\\\/d9569fde6c701b021a8d958f775be9a0\",\"name\":\"Dr. Duru\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2018\\\/09\\\/cropped-masthead-header.jpg\",\"url\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2018\\\/09\\\/cropped-masthead-header.jpg\",\"contentUrl\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2018\\\/09\\\/cropped-masthead-header.jpg\",\"width\":900,\"height\":100,\"caption\":\"Dr. Duru\"},\"logo\":{\"@id\":\"https:\\\/\\\/drduru.com\\\/onetwentytwo\\\/wp-content\\\/uploads\\\/2018\\\/09\\\/cropped-masthead-header.jpg\"},\"sameAs\":[\"http:\\\/\\\/www.drduru.com\\\/onetwentytwo\",\"https:\\\/\\\/www.instagram.com\\\/drduru_market_breadth\\\/\",\"https:\\\/\\\/x.com\\\/drduru\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"How to Use ChatGPT to Test Theories on Stock Prices","description":"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.","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:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/","og_locale":"en_US","og_type":"article","og_title":"How to Use ChatGPT to Test Theories on Stock Prices","og_description":"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.","og_url":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/","og_site_name":"ONE-TWENTY TWO: Trading Financial Markets","article_published_time":"2024-02-09T19:57:06+00:00","article_modified_time":"2024-02-11T04:41:53+00:00","og_image":[{"width":806,"height":495,"url":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png","type":"image\/png"}],"author":"Dr. Duru","twitter_card":"summary_large_image","twitter_title":"How to Use ChatGPT to Test Theories on Stock Prices","twitter_description":"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.","twitter_image":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png","twitter_creator":"@drduru","twitter_site":"@drduru","twitter_misc":{"Written by":"Dr. Duru","Est. reading time":"9 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#article","isPartOf":{"@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/"},"author":{"name":"Dr. Duru","@id":"https:\/\/drduru.com\/onetwentytwo\/#\/schema\/person\/d9569fde6c701b021a8d958f775be9a0"},"headline":"How to Use ChatGPT to Test Theories on Stock Prices","datePublished":"2024-02-09T19:57:06+00:00","dateModified":"2024-02-11T04:41:53+00:00","mainEntityOfPage":{"@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/"},"wordCount":1915,"commentCount":8,"publisher":{"@id":"https:\/\/drduru.com\/onetwentytwo\/#\/schema\/person\/d9569fde6c701b021a8d958f775be9a0"},"image":{"@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#primaryimage"},"thumbnailUrl":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png","keywords":["ChatGPT","data analyst","generative AI","Invesco QQQ Trust","iShares Russell 2000 ETF","IWM","prompt engineering","QQQ","statistical significance"],"articleSection":["Artificial Intelligence","Stock Market Trading","Technical Analysis","Trading Model","Trading Range"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/","url":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/","name":"How to Use ChatGPT to Test Theories on Stock Prices","isPartOf":{"@id":"https:\/\/drduru.com\/onetwentytwo\/#website"},"primaryImageOfPage":{"@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#primaryimage"},"image":{"@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#primaryimage"},"thumbnailUrl":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png","datePublished":"2024-02-09T19:57:06+00:00","dateModified":"2024-02-11T04:41:53+00:00","description":"Here is an example of the potential of ChatGPT for navigating the price dynamics in the stock market and to test theories on stock prices.","breadcrumb":{"@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#primaryimage","url":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png","contentUrl":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2024\/02\/20240209_IWM-vs-QQQ-Analysis-all-trading-data.png","width":806,"height":495,"caption":"Percentage of days IWM experiences a gain based on the combined performance of QQQ and IWM (February 12, 2019 to February 8, 2024)"},{"@type":"BreadcrumbList","@id":"https:\/\/drduru.com\/onetwentytwo\/2024\/02\/09\/how-to-use-chatgpt-to-test-theories-on-stock-prices\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/drduru.com\/onetwentytwo\/"},{"@type":"ListItem","position":2,"name":"How to Use ChatGPT to Test Theories on Stock Prices"}]},{"@type":"WebSite","@id":"https:\/\/drduru.com\/onetwentytwo\/#website","url":"https:\/\/drduru.com\/onetwentytwo\/","name":"ONE-TWENTY TWO: Trading Financial Markets","description":"Exploring the poetry in financial markets for students of money and economy since the year 2000.","publisher":{"@id":"https:\/\/drduru.com\/onetwentytwo\/#\/schema\/person\/d9569fde6c701b021a8d958f775be9a0"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/drduru.com\/onetwentytwo\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":["Person","Organization"],"@id":"https:\/\/drduru.com\/onetwentytwo\/#\/schema\/person\/d9569fde6c701b021a8d958f775be9a0","name":"Dr. Duru","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2018\/09\/cropped-masthead-header.jpg","url":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2018\/09\/cropped-masthead-header.jpg","contentUrl":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2018\/09\/cropped-masthead-header.jpg","width":900,"height":100,"caption":"Dr. Duru"},"logo":{"@id":"https:\/\/drduru.com\/onetwentytwo\/wp-content\/uploads\/2018\/09\/cropped-masthead-header.jpg"},"sameAs":["http:\/\/www.drduru.com\/onetwentytwo","https:\/\/www.instagram.com\/drduru_market_breadth\/","https:\/\/x.com\/drduru"]}]}},"_links":{"self":[{"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/posts\/67535","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/comments?post=67535"}],"version-history":[{"count":17,"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/posts\/67535\/revisions"}],"predecessor-version":[{"id":67556,"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/posts\/67535\/revisions\/67556"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/media\/67545"}],"wp:attachment":[{"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/media?parent=67535"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/categories?post=67535"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/drduru.com\/onetwentytwo\/wp-json\/wp\/v2\/tags?post=67535"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}