Asset manager builds daily country ETF map with ChatGPT

Andrew Rice, partner and portfolio manager at Algorithmic Investment Models, described using ChatGPT to build a daily country ETF performance map in the May 2026 edition of AIM’s Un-Herd newsletter. The tool compares 1-month trailing returns for 44 countries’ US-listed ETFs against the all-world index and the script can be run each morning.

Rice said he began by asking the language model to check his list of single-country ETFs. The model suggested three additional countries; those tickers were later confirmed to have been delisted. He settled on a final list of 44 countries with dedicated US-listed ETFs and asked for Python code to generate a daily map of each ETF’s trailing one-month return relative to the all-world index.

According to Rice, the model produced semi-workable code in roughly 90 seconds. He spent about 30–60 minutes of active time debugging and tweaking the output. When errors appeared, he pasted error messages into the chat and received corrected code back within 30–120 seconds of processing time. Rice estimated that resolving each bug alone could have taken one to four hours if done manually.

Rice included one line capturing the time savings: “I no longer needed to consider whether making that map was worth a week of my time because I could now accomplish it with about 30–60 minutes of active time.” He framed the example as a research workflow demonstration rather than an investment recommendation.

The newsletter notes that the project used third-party data and that the post is informational, not investment advice. The charts and map are based on external sources believed to be accurate, but no representation is made as to their completeness or future performance.

The work was published as part of AIM’s monthly newsletter. The write-up reflects an individual practitioner’s internal research process and is not presented as an AIM product recommendation.