{
  "annotation_id": "d69d6975-e0a9-49db-a577-16493416102a",
  "date_created": "2026-02-17T05:53:15.697000Z",
  "date_modified": "2026-02-17T05:53:15.697000Z",
  "file_hash": "1c2ad4d7b9dfc30f40fa57abbf043d657969010b86c1d0110682b45caef05dc0",
  "private": false,
  "record": {
    "abstract": "Large language models (LLMs) are a type of machine learning tool that economists have started to apply in their empirical research. One such application is macroeconomic forecasting with backtesting of LLMs, even though they are trained on the same data that is used to estimate their forecasting performance. Can these in-sample accuracy results be extrapolated to the model\u0027s out-of-sample performance? To answer this question, we developed a family of prompt sensitivity tests and two members of this family, which we call the fake date tests. These tests aim to detect two types of biases in LLMs\u0027 in-sample forecasts: lookahead bias and context bias. According to the empirical results, none of the modern LLMs tested in this study passed our first test, signaling the presence of lookahead bias in their in-sample forecasts.",
    "arxiv_id": "2601.07992",
    "authors": [
      "Alexander Eliseev",
      "Sergei Seleznev"
    ],
    "categories": [
      "econ.EM"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "title": "Fake Date Tests: Can We Trust In-sample Accuracy of LLMs in Macroeconomic Forecasting?",
    "url": "https://arxiv.org/abs/2601.07992",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "f448e7fa-4c2f-4be3-9ce3-691a6eabc65c",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}