LLM annotations of 1.28 million Korean financial texts (news headlines and
retail investor community posts) and the resulting daily fear–greed index for
the Korean stock market (research period 2021-01 – 2026-08, with a 2019–2020
news-only warm-up segment).
Companion dataset for: "Measuring fear and greed with large language models:
Evidence from news and retail investor communities in the Korean stock market".… See the full description on the dataset page:
https://huggingface.co/datasets/taejoon89/kor-fear-greed.
