关于「基础率与维度:凭单一行为贴标签的误判率」,《远期承诺期权与对等停时》给出的定量结论是什么?
结论:第三章说明,在基础率 0.16 下凭单一行为贴标签,错误率为 0.686。
数据口径:书中参数为示意;全部计算由书中程序生成;本书不是心理咨询、婚恋咨询或临床判断,也不做任何命理推导。 书中以标注框区分证据、可运行计算(21 段程序)、原典与作者假说。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 与结论吻合的打印值: 0.16, 0.686
| sensitivity 0.60, false-positive rate 0.25, LR = 2.40 | |||
| base rate | P(high|seen) | wrong labels | labels per true case |
| 0.05 | 0.112 | 0.888 | 8.92 |
| 0.10 | 0.211 | 0.789 | 4.75 |
| 0.16 | 0.314 | 0.686 | 3.19 |
| 0.25 | 0.444 | 0.556 | 2.25 |
| 0.40 | 0.615 | 0.385 | 1.63 |
| 0.50 | 0.706 | 0.294 | 1.42 |
| LR needed for P=0.90 at base rate 0.16: 47.3 |
查看完整验算程序
# Chapter 3. p31: labelling from one behaviour, error set by base rate
# A single observed behaviour (say, steering away from a dated plan)
# appears with probability SENS when the other party sits high on the
# avoidance dimension and FPR otherwise (busy weeks, other reasons).
# Bayes: P(high | behaviour) = SENS*pi / (SENS*pi + FPR*(1-pi)).
# Bar-Hillel (1980) on neglect of pi. Numbers illustrative.
SENS, FPR = 0.60, 0.25
print("sensitivity 0.60, false-positive rate 0.25, LR = {:.2f}".format(
SENS / FPR))
print("base rate P(high|seen) wrong labels labels per true case")
for pi in (0.05, 0.10, 0.16, 0.25, 0.40, 0.50):
hit = SENS * pi
fa = FPR * (1.0 - pi)
ppv = hit / (hit + fa)
per_true = (hit + fa) / hit
print("{:9.2f} {:14.3f} {:13.3f} {:20.2f}".format(
pi, ppv, 1.0 - ppv, per_true))
need = 0.90
pi = 0.16
lr_need = need / (1.0 - need) / (pi / (1.0 - pi))
print("LR needed for P=0.90 at base rate 0.16: {:.1f}".format(lr_need))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《亲职基因优选与排他恋爱解耦、远期空头期权钩子与对等底线停时母典》第 3 章「基础率与维度:凭单一行为贴标签的误判率」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-423 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026423,
author = {Fan, Yafu},
title = {Unbundling Paternal Selection from Romantic Commitment: Deferred-Promise Options, Avoidant Autonomy and Reciprocal Stopping Rules},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}