关于「前后对比的三重混杂:季节、大盘与噪声怎样冒充改版效应」,《合成控制与单站改版》给出的定量结论是什么?
结论:前后对比的平均估计是 45.64,偏差 33.64,均方根误差 35.61;两百个面板里有 0.980 的比例误差超过 6 单,有 0.970 的比例报出了真实提升两倍以上的数。这说明在这样的改版时点,前后对比几乎每一次都会把改版吹得很大。双重差分的偏差降到 10.26,标准差只有 3.75,均方根误差 10.92;它稳定,但稳定地偏,0.320 的面板报出两倍以上的提升。合成对照的平均估计是 15.90,偏差 3.90,标准差 3.77,均方根误差 5.42;超过真实值两倍的只有 0.015。按均方根误差算,前后对比是合成对照的 6.56 倍,双重差分是 2.01 倍。
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: 0.015, 0.320, 0.970, 0.980, 10.26, 10.92
| 200 panels, true lift +12.0 orders/week | ||||||
| method | mean | bias | sd | RMSE | |err|>6 | est>24 |
| before-after | 45.64 | +33.64 | 11.69 | 35.61 | 0.980 | 0.970 |
| diff-in-diff | 22.26 | +10.26 | 3.75 | 10.92 | 0.835 | 0.320 |
| synthetic | 15.90 | +3.90 | 3.77 | 5.42 | 0.275 | 0.015 |
| RMSE ratio vs synthetic: before-after 6.56, DiD 2.01 |
查看完整验算程序
# Chapter 1. p12: 200 redrawn panels: bias, spread and big misses.
# Model as in p11 (illustrative factor model, 34 sites, 52 + 12
# weeks, true lift +12). Each replicate redraws donors, market path
# and noise with a fixed seed sequence; the redesigned site keeps the
# same loadings. We report the mean estimate, bias, standard
# deviation, RMSE and the share of replicates whose error exceeds
# half the true lift (6 orders/week) or that double the lift (> 24).
# 参数为示意;固定种子,逐字可复现。
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__":
reps = 200
pre12, post = list(range(T0 - 12, T0)), list(range(T0, T0 + POST))
donors = list(range(1, NS))
est = {"before-after": [], "diff-in-diff": [], "synthetic": []}
for r in range(reps):
ys, cf, pars, ss, mk = panel(r)
ch = [mean([y[t] for t in post]) - mean([y[t] for t in pre12])
for y in ys]
est["before-after"].append(ch[0])
est["diff-in-diff"].append(ch[0] - mean(ch[1:]))
g = gram(ys, T0)
est["synthetic"].append(mean(synth(ys, g, 0, donors)[1][T0:]))
print("{} panels, true lift +12.0 orders/week".format(reps))
print("method mean bias sd RMSE |err|>6 est>24")
for k, xs in est.items():
m = mean(xs)
sd = math.sqrt(math.fsum([(x - m) ** 2 for x in xs]) / reps)
rm = math.sqrt(math.fsum([(x - 12.0) ** 2 for x in xs]) / reps)
big = len([x for x in xs if abs(x - 12.0) > 6.0]) / reps
dbl = len([x for x in xs if x > 24.0]) / reps
print("{:<13s} {:6.2f} {:+7.2f} {:6.2f} {:6.2f} {:.3f}"
" {:.3f}".format(k, m, m - 12.0, sd, rm, big, dbl))
ratio_rm = []
for k in ("before-after", "diff-in-diff"):
a = math.fsum([(x - 12.0) ** 2 for x in est[k]])
b = math.fsum([(x - 12.0) ** 2 for x in est["synthetic"]])
ratio_rm.append(math.sqrt(a / b))
print("RMSE ratio vs synthetic: before-after {:.2f}, DiD {:.2f}"
.format(ratio_rm[0], ratio_rm[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 1 章「前后对比的三重混杂:季节、大盘与噪声怎样冒充改版效应」(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}
}