关于「动态模态分解:特征值、振幅与噪声造成的假衰减」,《库普曼模态与直播拐点》给出的定量结论是什么?

结论:第四章组装动态模态分解,指出噪声会把永不衰减的起伏估成模长 0.7744、半衰期 2.7 分钟,前向后向修正可以纠正。

数据口径:书中的直播曲线、店铺流水与参数均为示意;全部计算由书中程序生成;本书是非个人化经营建议,现金流预报不构成融资、借贷或投资建议,书中小工具给出的每一条预警都只是内部提示,须由经营者人工判断。 第一章显示示意直播曲线的前两个模态装下 0.9285 的能量,但随机游走也有 0.872,所以方差解释不是证据,样本外预报才是;去掉 12 分钟起伏的移动平均,拐点迟到 5.89 分钟。

English version

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: 0.7744, 2.7

rowssamplesnoiseDMD |lambda|half-lifefbDMD |lambda|
424600.980935.9 min1.0000
4241500.90036.6 min1.0012
4243000.77442.7 min1.0065
1030600.9992852.5 min1.0013
10301500.985146.1 min0.9965
10303000.946312.6 min0.9969
2060600.9993974.4 min0.9997
20601500.9960172.8 min0.9987
20603000.984343.7 min0.9963
medians over 100 replicates; true |lambda| = 1, half-life inf
查看完整验算程序
# Chapter 4. p42: sensor noise makes a steady cycle look as if it dies.
# Signal: a 12-minute cycle of amplitude 300 that never decays (true
# |lambda| = 1), plus Gaussian noise; 100 fixed-seed replicates per cell.
# Standard DMD (rank 2) is compared with forward-backward DMD (Dawson et
# al. 2016): fit the forward map A_f and the backward map A_b in the same
# POD basis and take eigenvalues of (A_f A_b^-1)^(1/2). "Half-life" is
# the apparent time for the cycle to halve, ln 2 / -ln |lambda|.


import math
import cmath
import random
import statistics


def dot(a, b):
    return math.fsum(x * y for x, y in zip(a, b))


def matmul(a, b):
    bt = list(zip(*b))
    return [[dot(r, c) for c in bt] for r in a]


def transpose(a):
    return [list(r) for r in zip(*a)]


def svd(a):
    # 单边 Jacobi(Hestenes):两两旋转列直到正交,A = U S V^T
    m, n = len(a), len(a[0])
    if n > m:
        u, s, v = svd(transpose(a))
        return v, s, u
    cols = [[a[i][j] for i in range(m)] for j in range(n)]
    vc = [[1.0 if i == j else 0.0 for i in range(n)] for j in range(n)]
    for sweep in range(60):
        rotated = False
        for p in range(n - 1):
            for q in range(p + 1, n):
                al = dot(cols[p], cols[p])
                be = dot(cols[q], cols[q])
                ga = dot(cols[p], cols[q])
                if ga == 0.0 or abs(ga) <= 1e-15 * math.sqrt(al * be):
                    continue
                rotated = True
                ze = (be - al) / (2.0 * ga)
                t = math.copysign(1.0, ze) / (
                    abs(ze) + math.sqrt(1.0 + ze * ze))
                c = 1.0 / math.sqrt(1.0 + t * t)
                s = c * t
                cp, cq = cols[p], cols[q]
                cols[p] = [c * x - s * y for x, y in zip(cp, cq)]
                cols[q] = [s * x + c * y for x, y in zip(cp, cq)]
                vp, vq = vc[p], vc[q]
                vc[p] = [c * x - s * y for x, y in zip(vp, vq)]
                vc[q] = [s * x + c * y for x, y in zip(vp, vq)]
        if not rotated:
            break
    sig = [math.sqrt(dot(c_, c_)) for c_ in cols]
    order = sorted(range(n), key=lambda j: (-sig[j], j))
    s = [sig[j] for j in order]
    u = [[cols[j][i] / sig[j] if sig[j] > 0.0 else 0.0 for j in order]
         for i in range(m)]
    v = [[vc[j][i] for j in order] for i in range(n)]
    return u, s, v


def hankel(x, rows):
    # 延迟嵌入:第 i 行是从第 i 分钟起的一段序列
    k = len(x) - rows + 1
    return [[x[i + j] for j in range(k)] for i in range(rows)]


def inv(a):
    # Gauss-Jordan 求逆(部分主元),只用于小矩阵
    n = len(a)
    m = [list(map(float, a[i])) + [1.0 if i == j else 0.0
                                   for j in range(n)] for i in range(n)]
    for c in range(n):
        p = max(range(c, n), key=lambda i: abs(m[i][c]))
        m[c], m[p] = m[p], m[c]
        pv = m[c][c]
        m[c] = [q / pv for q in m[c]]
        for i in range(n):
            if i != c:
                f = m[i][c]
                m[i] = [q - f * w for q, w in zip(m[i], m[c])]
    return [row[n:] for row in m]


def charpoly(a):
    # Faddeev-LeVerrier:特征多项式 z^n + c1 z^(n-1) + c2 z^(n-2) 直到 cn
    n = len(a)
    m = [[0.0] * n for _ in range(n)]
    coef = [1.0]
    for k in range(1, n + 1):
        m = matmul(a, m)
        for i in range(n):
            m[i][i] += coef[-1]
        am = matmul(a, m)
        coef.append(-math.fsum(am[i][i] for i in range(n)) / k)
    return coef


def roots(coef):
    # Durand-Kerner 同时迭代求全部复根(cmath)
    n = len(coef) - 1
    z = [cmath.rect(0.9, 0.4 + 2.0 * math.pi * k / n) for k in range(n)]

    def p(x):
        r = complex(1.0, 0.0)
        for c in coef[1:]:
            r = r * x + c
        return r
    for _ in range(800):
        new = []
        for i in range(n):
            den = complex(1.0, 0.0)
            for j in range(n):
                if j != i:
                    den *= z[i] - z[j]
            new.append(z[i] - p(z[i]) / den)
        moved = max(abs(a - b) for a, b in zip(new, z))
        z = new
        if moved < 1e-14:
            break
    return z


def eigvals(a):
    z = roots(charpoly(a))
    z = [complex(w.real, 0.0) if abs(w.imag) < 1e-12 else w for w in z]
    return sorted(z, key=lambda w: (-round(abs(w), 9), w.imag))


def pinv_rows(xr):
    return matmul(transpose(xr), inv(matmul(xr, transpose(xr))))


def moduli(x, rows, r):
    h = hankel(x, rows)
    xs = [row[:-1] for row in h]
    ys = [row[1:] for row in h]
    u, s, v = svd(xs)
    ur = transpose([row[:r] for row in u])
    xr, yr = matmul(ur, xs), matmul(ur, ys)
    af = matmul(yr, pinv_rows(xr))
    ab = matmul(xr, pinv_rows(yr))
    lf = eigvals(af)
    lfb = [cmath.sqrt(z) for z in eigvals(matmul(af, inv(ab)))]
    return abs(lf[0]), abs(lfb[0])


rng = random.Random(20261116)
print("rows  samples  noise   DMD |lambda|  half-life   fbDMD |lambda|")
for rows, n in ((4, 24), (10, 30), (20, 60)):
    for sd in (60.0, 150.0, 300.0):
        a, b = [], []
        for _ in range(100):
            x = [300.0 * math.cos(2.0 * math.pi * t / 12.0)
                 + rng.gauss(0.0, sd) for t in range(n)]
            f, fb = moduli(x, rows, 2)
            a.append(f)
            b.append(fb)
        ma, mb = statistics.median(a), statistics.median(b)
        hl = math.log(2.0) / -math.log(ma) if ma < 1.0 else float("inf")
        print("%4d  %7d  %5.0f   %12.4f  %6.1f min   %14.4f"
              % (rows, n, sd, ma, hl, mb))
print("medians over 100 replicates; true |lambda| = 1, half-life inf")

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《库普曼算子动态模态分解、现金流非线性波动线性化与直播在线曲线拐点预测母典》第 4 章「动态模态分解:特征值、振幅与噪声造成的假衰减」(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-238 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026238,
  author    = {Fan, Yafu},
  title     = {Koopman Operators and Dynamic Mode Decomposition: Linearising Nonlinear Cash-Flow Swings and Forecasting Turning Points in Livestream Audience Curves},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论