What does Synthetic Control for Site Redesigns conclude, in numbers, about “Simplex-Constrained Least Squares: How Donor Weights Are Found and Why They May Not Be Negative”?
Answer: Chapter 2 finds simplex-constrained least-squares weights by projected gradient and by exact enumeration, and shows that with only thirty-six pre weeks the error of free regression rises to 18.22 while simplex weights stay at 5.34.
Scope of the numbers: On an illustrative panel of thirty-four sites with a true lift of twelve orders a week, before-after reports 37.77, mostly because of the season; weighting thirty-three untouched donor sites with simplex-constrained weights into a synthetic site gives 8.83. Over two hundred panels the root mean squared error is 35.61 for before-after and 5.42 for synthetic control.
Data table
The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 18.22, 5.34
one panel preRMSPE postCF-RMSE estimate sum|w| neg max w free OLS 4.24 9.44 10.46 4.58 17 0.33 affine 4.25 9.54 10.57 4.58 17 0.34 simplex 5.90 5.31 8.83 1.00 0 0.32 200 panels: RMSE of lift / mean weekly post-CF RMSE pre weeks 52: OLS 5.60/12.53 affine 5.44/12.15 simplex 5.42/8.90 pre weeks 40: OLS 8.85/18.51 affine 8.19/17.31 simplex 5.43/8.88 pre weeks 36: OLS 18.22/28.27 affine 12.42/24.03 simplex 5.34/9.03
Show the full program
# Chapter 2. p23: why the simplex: free regression and affine weights.
# Three ways to weight the 33 donors on the 52 pre weeks:
# free OLS - any real weights, no intercept (normal equations);
# affine - weights sum to one, negatives allowed (KKT system);
# simplex - weights >= 0 and sum to one (synthetic control).
# For one panel we print pre RMSPE, RMSE of the counterfactual over the
# 12 post weeks (against the known no-redesign path), the estimate and
# the weight shape; then, over 200 panels with 52, 40 or 36 pre weeks,
# the RMSE of the estimated lift and the mean weekly post-CF RMSE.
# Panel as in p11 (illustrative; true lift +12).
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
def fit(ys, g, method, donors, t0=T0):
gs = [[g[i][k] for k in donors] for i in donors]
cs = [g[i][0] for i in donors]
if method == "free OLS":
w = solve(gs, cs)
elif method == "affine":
w = eq_qp(gs, cs, list(range(len(donors))))
else:
w = simplex_ls(gs, cs)
yh = [dot(w, [ys[d][t] for d in donors]) for t in range(t0 + POST)]
return w, yh
if __name__ == "__main__":
donors = list(range(1, NS))
meths = ("free OLS", "affine", "simplex")
ys, cf, pars, ss, mk = panel(0)
g = gram(ys, T0)
print("one panel preRMSPE postCF-RMSE estimate sum|w| neg"
" max w")
for m in meths:
w, yh = fit(ys, g, m, donors)
pre = rmspe([ys[0][t] - yh[t] for t in range(T0)])
pcf = rmspe([cf[0][t] - yh[t] for t in range(T0, T0 + POST)])
est = mean([ys[0][t] - yh[t] for t in range(T0, T0 + POST)])
print("{:<12s} {:8.2f} {:10.2f} {:9.2f} {:6.2f} {:3d} {:5.2f}"
.format(m, pre, pcf, est, math.fsum([abs(x) for x in w]),
len([x for x in w if x < 0.0]), max(w)))
print("200 panels: RMSE of lift / mean weekly post-CF RMSE")
for t0 in (52, 40, 36):
errs = {m: [] for m in meths}
wk = {m: [] for m in meths}
for r in range(200):
ys, cf, pars, ss, mk = panel(r, t0=t0, start=(62 - t0) % 52)
g = gram(ys, t0)
for m in meths:
w, yh = fit(ys, g, m, donors, t0)
po = range(t0, t0 + POST)
est = mean([ys[0][t] - yh[t] for t in po])
errs[m].append(est - 12.0)
wk[m].append(rmspe([cf[0][t] - yh[t] for t in po]))
print(" pre weeks {:2d}: ".format(t0) + " ".join(
"{} {:.2f}/{:.2f}".format(m.split()[-1], rmspe(errs[m]),
mean(wk[m])) for m in meths))Single source and citation
Fan, Y. (2026). Synthetic Control and the Causal Effect of a Single-Site Redesign: Rebuilding the Counterfactual from Weighted Donor Sites and Placebo Permutation Tests (Version 1.0.0), Chapter 2: Simplex-Constrained Least Squares: How Donor Weights Are Found and Why They May Not Be Negative. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 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}
}Other chapter answers in this book
- Chapter 1 · Three Confounders in a Before-After Comparison: How Season, Market and Noise Pose as a Redesign Effect
- Chapter 3 · Pre-Period Fit Is the Entry Ticket: Root Mean Squared Prediction Error, the Convex Hull and Hold-Out Checks
- Chapter 4 · The Placebo Permutation Test: the Post/Pre Ratio, Ranks and the Floor of the p-Value
- Chapter 5 · The Donor Pool: No Spillover, Enough Donors and Interpolation Bias
- Chapter 6 · Power and the Length of the Evaluation: How Many Weeks to See Five Per Cent
- Chapter 7 · Eastern and Western Sources: "Take by Kind, Give by Kind" in the Mozi and Mill's Method of Difference
- Chapter 8 · Closing the Loop: Evaluating Single-Site Redesigns across Thirty-Four Sites, with a Pre-Registered Protocol