跳至主要內容

文章

NeuPortal 量化實驗室

關於 AI 代理、回測與市場預測稽核,以及平台架構的研究。

預測評分 · 13

  1. 預測評分

    AI and Calibration: Why a Good 70% Forecast Is Wrong Three Times in Ten

    The instinct to grade a forecast by whether it came true is the wrong test. Calibration - when a model says 70% it happens about 70% of the time - is the property that separates a real forecaster from a confident guesser, and it is measurable.

    閱讀時間 7 分鐘7 個章節

  2. 預測評分

    What Is the CRPS? Grading a Whole Probability Distribution

    The Brier score grades a probability for a yes/no event. The CRPS does the same job for a full distribution over a number. Here is how it works and why it is a proper scoring rule.

    閱讀時間 3 分鐘5 個章節

  3. 預測評分

    How to Compare Forecasting Models Fairly

    A rigorous framework for comparing forecasting models without bias — covering evaluation protocols, metrics, baselines, and statistical significance tests.

    閱讀時間 7 分鐘9 個章節

我們使用 Cookie 以確保 AI 交易終端與支付閘道正常運作,並完成身分驗證。繼續瀏覽本網站,即表示您同意我們的Cookie 政策及隱私權政策。

選擇「接受並繼續」即表示您同時允許使用 Google Ads 廣告 Cookie,用於衡量我們的廣告成效。這些 Cookie 絕不會在使用者帳戶中使用。