解释概率预测的相对效用
Interpreting relative utility for probabilistic predictions
浏览论文内容
中文总结 AI 辅助
本文阐明相对效用（RU）在固定阈值下衡量的是完美结局分类而非完美概率预测，并指出即使预测概率完全正确，观测RU也可能为0，因此RU与1的距离不应被解释为预测改进空间。
中文摘要 AI 辅助
在固定阈值下，相对效用（RU）衡量预测模型相对于治疗所有患者和治疗无患者中较优者的净收益增益，该增益以完美结局分类下的相应增益为基准。我们说明，RU等于1因此代表在该固定阈值下的完美结局分类，而非完美的概率预测。然而，即使每个预测概率都等于真实概率，观测到的RU也可能等于0。在一个简单的恒定风险设置中，随着样本量增加，这种情况发生的概率趋近于1。因此，观测RU与1之间的距离通常不应被解释为通过更好的二元概率预测所能实现的改进。
英文摘要
At a fixed threshold, relative utility (RU) measures the net-benefit gain of a prediction model over the better of treat-all and treat-none relative to the corresponding gain under perfect outcome classification. We illustrate that RU equal to 1 therefore represents perfect outcome classification at that fixed threshold, not perfect probabilistic prediction. However, even when every predicted probability equals the true probability, observed RU can equal 0. In a simple constant-risk setting, this occurs with probability approaching 1 as the sample size increases. Consequently, the distance from observed RU to 1 should not in general be interpreted as improvement achievable by a better prediction for binary probabilities.
发表机构
- Data Center of the Swiss Transplant Cohort Study, University of Basel & University Hospital Basel（瑞士移植队列研究数据中心，巴塞尔大学与巴塞尔大学医院）
机构由 AI 辅助整理，请以论文原文为准。
