关于「捐赠池:无溢出、捐赠站数与插值偏差」,《合成控制与单站改版》给出的定量结论是什么?
结论:只有三个捐赠站时,前期误差平均 19.95,0.745 的面板拟合不及格,提升估计的均方根误差高达 31.81,比前后对比好不了多少;p 值的地板是 0.2500,无论效应多大都不可能显著。五个站时拟合不及格的比例仍有 0.495,地板 0.1667。十个站时拟合不及格降到 0.125,提升误差 9.47,地板 0.0909,仍然够不着百分之五。二十个站时,前期误差 7.52,不及格 0.030,提升误差 6.78,地板 0.0476,第一次跨过百分之五的门槛。三十三个站时,前期误差 6.84,不及格 0.000,提升误差 5.42,地板 0.0294。
数据口径:在一个三十四站的示意面板上,真实提升设为每周十二单,前后对比报出 37.77,主要误差来自季节;用单纯形约束权重把三十三个未改版的捐赠站加权成合成对照站,估计为 8.83。 两百个面板上,前后对比的均方根误差为 35.61,合成对照为 5.42。
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。 程序只用 Python 标准库与结论吻合的打印值: 0.000, 0.0294, 0.030, 0.0476, 0.0909, 0.125
| J | preRMSPE | fit>10.5 | lift RMSE | min p |
| 3 | 19.95 | 0.745 | 31.81 | 0.2500 |
| 5 | 12.47 | 0.495 | 19.33 | 0.1667 |
| 10 | 8.66 | 0.125 | 9.47 | 0.0909 |
| 20 | 7.52 | 0.030 | 6.78 | 0.0476 |
| 33 | 6.84 | 0.000 | 5.42 | 0.0294 |
查看完整验算程序
# Chapter 5. p52: how many donors: fit quality versus the p-value floor.
# Panel as in p11 (illustrative, true lift +12). From the 33 untreated
# sites a random pool of J donors is drawn (fixed seed) for each of
# 200 panels; J = 3, 5, 10, 20, 33. Reported: mean pre RMSPE, share of
# panels whose pre RMSPE exceeds 1.5 times the site's noise sd (10.5),
# RMSE of the estimated lift, and the smallest attainable placebo
# p-value 1/(J + 1).
import math
import random
from operator import mul
# 示意面板:每站周订单 = 基准 + 季节载荷 * 季节 + 大盘载荷 * 大盘 + 噪声
T0, POST, NS = 52, 12, 34
def panel(rep=0, n=NS, t0=T0, post=POST, tau=12.0, start=10):
rng = random.Random(20261113 + 7919 * rep)
weeks = t0 + post
mk = [0.0]
for _ in range(weeks - 1):
mk.append(mk[-1] + rng.gauss(0.05, 0.25))
ss = [math.sin(2.0 * math.pi * (start + t) / 52.0)
for t in range(weeks)]
ys, cf, pars = [], [], []
for i in range(n):
if i == 0:
mu, ra, rb, sd = 240.0, 0.20, 1.30, 7.0
else:
mu = rng.uniform(120.0, 360.0)
ra = rng.uniform(0.04, 0.22)
rb = rng.uniform(0.5, 1.7)
sd = rng.uniform(5.0, 9.0)
a, b = mu * ra, mu * rb / 20.0
y = [mu + a * ss[t] + b * mk[t] + rng.gauss(0.0, sd)
for t in range(weeks)]
cf.append(y[:])
if i == 0:
for t in range(t0, weeks):
y[t] = y[t] + tau
ys.append(y)
pars.append((mu, a, b, sd))
return ys, cf, pars, ss, mk
def dot(a, b):
return math.fsum(map(mul, a, b))
def mean(xs):
return math.fsum(xs) / len(xs)
def rmspe(xs):
return math.sqrt(math.fsum([x * x for x in xs]) / len(xs))
def gram(ys, t0):
n = len(ys)
pre = [y[:t0] for y in ys]
g = [[0.0] * n for _ in range(n)]
for i in range(n):
for k in range(i, n):
v = dot(pre[i], pre[k]) / t0
g[i][k] = v
g[k][i] = v
return g
def solve(a, b):
m = len(b)
r = [row[:] + [b[i]] for i, row in enumerate(a)]
for c in range(m):
p = max(range(c, m), key=lambda i: abs(r[i][c]))
r[c], r[p] = r[p], r[c]
for i in range(c + 1, m):
f = r[i][c] / r[c][c]
if f:
r[i] = [x - f * y for x, y in zip(r[i], r[c])]
x = [0.0] * m
for i in range(m - 1, -1, -1):
s = math.fsum([r[i][k] * x[k] for k in range(i + 1, m)])
x[i] = (r[i][m] - s) / r[i][i]
return x
def eq_qp(g, c, sup):
m = len(sup)
a = [[2.0 * g[i][k] for k in sup] + [1.0] for i in sup]
a.append([1.0] * m + [0.0])
b = [2.0 * c[i] for i in sup] + [1.0]
return solve(a, b)[:m]
def simplex_ls(g, c, tol=1e-10):
# 精确有效集法:min w'Gw - 2c'w,w >= 0,权重和为 1
j = len(c)
k0 = min(range(j), key=lambda k: g[k][k] - 2.0 * c[k])
w = [0.0] * j
w[k0] = 1.0
sup = [k0]
for _ in range(4 * j + 20):
gr = [2.0 * (math.fsum([g[i][k] * w[k] for k in sup]) - c[i])
for i in range(j)]
nu = -math.fsum([gr[k] for k in sup]) / len(sup)
out = [k for k in range(j) if k not in sup]
if not out:
break
kb = min(out, key=lambda k: gr[k] + nu)
if gr[kb] + nu >= -tol:
break
sup = sorted(sup + [kb])
while True:
z = eq_qp(g, c, sup)
if min(z) > 0.0:
for k, v in zip(sup, z):
w[k] = v
break
step = min(w[k] / (w[k] - v) for k, v in zip(sup, z)
if v <= 0.0)
for k, v in zip(sup, z):
w[k] = w[k] + step * (v - w[k])
sup = [k for k in sup if w[k] > 1e-12]
for k in range(j):
if k not in sup:
w[k] = 0.0
return w
def synth(ys, g, tr, donors):
gs = [[g[i][k] for k in donors] for i in donors]
cs = [g[i][tr] for i in donors]
w = simplex_ls(gs, cs)
weeks = len(ys[tr])
gap = [ys[tr][t] - dot(w, [ys[d][t] for d in donors])
for t in range(weeks)]
return w, gap
def ratio(gap, t0=T0):
return rmspe(gap[t0:]) / rmspe(gap[:t0])
def placebo(ys, g, tr, pool, t0=T0):
# 逐个捐赠站当处理站;处理站本身不进入安慰剂的捐赠池
out = {}
for j in pool:
dn = [k for k in pool if k != j]
out[j] = synth(ys, g, j, dn)[1]
return out
if __name__ == "__main__":
pick = random.Random(20261113)
print(" J preRMSPE fit>10.5 lift RMSE min p")
for jn in (3, 5, 10, 20, 33):
pre, err = [], []
for r in range(200):
ys, cf, pars, ss, mk = panel(r)
pool = sorted(pick.sample(range(1, NS), jn))
g = gram(ys, T0)
w, gap = synth(ys, g, 0, pool)
pre.append(rmspe(gap[:T0]))
err.append(mean(gap[T0:]) - 12.0)
bad = len([x for x in pre if x > 10.5]) / len(pre)
print("{:3d} {:8.2f} {:8.3f} {:9.2f} {:.4f}".format(
jn, mean(pre), bad, rmspe(err), 1 / (jn + 1)))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《合成控制法与单站改版因果、捐赠站加权重建反事实与安慰剂置换检验母典》第 5 章「捐赠池:无溢出、捐赠站数与插值偏差」(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-304 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026304,
author = {Fan, Yafu},
title = {Synthetic Control and the Causal Effect of a Single-Site Redesign: Rebuilding the Counterfactual from Weighted Donor Sites and Placebo Permutation Tests},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}