@Jean-Paul算法的Perl实现。

#!/usr/bin/perl

use strict;
use Data::Dumper;

sub mean {
    my $data = shift;
    my $sum = 0;
    my $mean_val = 0;
    for my $item (@$data) {
        $sum += $item;
    }
    $mean_val = $sum / (scalar @$data) if @$data;
    return $mean_val;
}

sub variance {
    my $data = shift;
    my $variance_val = 0;
    my $mean_val = mean($data);
    my $sum = 0;
    for my $item (@$data) {
        $sum += ($item - $mean_val)**2;
    }
    $variance_val = $sum / (scalar @$data) if @$data;
    return $variance_val;
}

sub std {
    my $data = shift;
    my $variance_val = variance($data);
    return sqrt($variance_val);
}

# @param y - The input vector to analyze
# @parameter lag - The lag of the moving window
# @parameter threshold - The z-score at which the algorithm signals
# @parameter influence - The influence (between 0 and 1) of new signals on the mean and standard deviation
sub thresholding_algo {
    my ($y, $lag, $threshold, $influence) = @_;

    my @signals = (0) x @$y;
    my @filteredY = @$y;
    my @avgFilter = (0) x @$y;
    my @stdFilter = (0) x @$y;

    $avgFilter[$lag - 1] = mean([@$y[0..$lag-1]]);
    $stdFilter[$lag - 1] = std([@$y[0..$lag-1]]);

    for (my $i=$lag; $i <= @$y - 1; $i++) {
        if (abs($y->[$i] - $avgFilter[$i-1]) > $threshold * $stdFilter[$i-1]) {
            if ($y->[$i] > $avgFilter[$i-1]) {
                $signals[$i] = 1;
            } else {
                $signals[$i] = -1;
            }

            $filteredY[$i] = $influence * $y->[$i] + (1 - $influence) * $filteredY[$i-1];
            $avgFilter[$i] = mean([@filteredY[($i-$lag)..($i-1)]]);
            $stdFilter[$i] = std([@filteredY[($i-$lag)..($i-1)]]);
        }
        else {
            $signals[$i] = 0;
            $filteredY[$i] = $y->[$i];
            $avgFilter[$i] = mean([@filteredY[($i-$lag)..($i-1)]]);
            $stdFilter[$i] = std([@filteredY[($i-$lag)..($i-1)]]);
        }
    }

    return {
        signals => \@signals,
        avgFilter => \@avgFilter,
        stdFilter => \@stdFilter
    };
}

my $y = [1,1,1.1,1,0.9,1,1,1.1,1,0.9,1,1.1,1,1,0.9,1,1,1.1,1,1,1,1,1.1,0.9,1,1.1,1,1,0.9,
       1,1.1,1,1,1.1,1,0.8,0.9,1,1.2,0.9,1,1,1.1,1.2,1,1.5,1,3,2,5,3,2,1,1,1,0.9,1,1,3,
       2.6,4,3,3.2,2,1,1,0.8,4,4,2,2.5,1,1,1];

my $lag = 30;
my $threshold = 5;
my $influence = 0;

my $result = thresholding_algo($y, $lag, $threshold, $influence);

print Dumper $result;

c++ (Qt)演示端口，交互式参数

我已经将这个算法的演示应用程序移植到c++ (Qt)上。

代码可以在GitHub上找到这里。带有安装程序的Windows(64位)构建在发布页面上。最后，我将添加一些文档和其他发布版本。

您不能绘制点，但可以从文本文件中导入它们(用空格分隔点——换行也算作空格)。您还可以调整算法参数，实时查看效果。这对于针对特定数据集调整算法以及探索参数如何影响结果非常有用。

上面的截图有些过时;从那以后，我添加了两个原始算法中没有的实验性选项:

反向处理数据集的选项(似乎至少改善了功率谱的结果)。 选项，为峰值设置硬性最小阈值。

我还在窗口中间添加了一个笨拙的缩放/平移条，只需用鼠标拖动它来缩放和平移。

模糊的构建指令:

在发布页面上有一个Windows安装程序(64位)，但如果你想从源代码构建它，要点是:

安装Qt的构建工具，然后将qmake && make放在与.pro文件相同的目录下，或者 安装Qt Creator，打开.pro文件，选择任何默认的构建配置，然后按下构建和/或运行按钮(Creator的左下角)。

我只测试过Qt5。我有91%的信心，如果你手动配置组件，Qt Creator安装程序会让你安装Qt5(如果你手动配置组件，你还需要确认是否安装了Qt Charts)。Qt6可能是一个流畅的构建，也可能不是。有一天，我将测试Qt4和Qt6，使这些文档更好。也许吧。

如果你的数据在一个数据库表中，这里是一个简单的z-score算法的SQL版本:

with data_with_zscore as (
    select
        date_time,
        value,
        value / (avg(value) over ()) as pct_of_mean,
        (value - avg(value) over ()) / (stdev(value) over ()) as z_score
    from {{tablename}}  where datetime > '2018-11-26' and datetime < '2018-12-03'
)


-- select all
select * from data_with_zscore 

-- select only points greater than a certain threshold
select * from data_with_zscore where z_score > abs(2)

我认为delica的Python回答器有一个bug。我不能评论他的帖子，因为我没有代表来做这件事，编辑队列已经满了，所以我可能不是第一个注意到它的人。

avgFilter[lag - 1]和stdFilter[lag - 1]在init中设置，然后在lag == i时再次设置，而不是改变[lag]值。这个结果使得第一个信号总是1。

以下是带有轻微修正的代码:

import numpy as np

class real_time_peak_detection():
    def __init__(self, array, lag, threshold, influence):
        self.y = list(array)
        self.length = len(self.y)
        self.lag = lag
        self.threshold = threshold
        self.influence = influence
        self.signals = [0] * len(self.y)
        self.filteredY = np.array(self.y).tolist()
        self.avgFilter = [0] * len(self.y)
        self.stdFilter = [0] * len(self.y)
        self.avgFilter[self.lag - 1] = np.mean(self.y[0:self.lag]).tolist()
        self.stdFilter[self.lag - 1] = np.std(self.y[0:self.lag]).tolist()

    def thresholding_algo(self, new_value):
        self.y.append(new_value)
        i = len(self.y) - 1
        self.length = len(self.y)
        if i < self.lag:
            return 0
        elif i == self.lag:
            self.signals = [0] * len(self.y)
            self.filteredY = np.array(self.y).tolist()
            self.avgFilter = [0] * len(self.y)
            self.stdFilter = [0] * len(self.y)
            self.avgFilter[self.lag] = np.mean(self.y[0:self.lag]).tolist()
            self.stdFilter[self.lag] = np.std(self.y[0:self.lag]).tolist()
            return 0

        self.signals += [0]
        self.filteredY += [0]
        self.avgFilter += [0]
        self.stdFilter += [0]

        if abs(self.y[i] - self.avgFilter[i - 1]) > self.threshold * self.stdFilter[i - 1]:
            if self.y[i] > self.avgFilter[i - 1]:
                self.signals[i] = 1
            else:
                self.signals[i] = -1

            self.filteredY[i] = self.influence * self.y[i] + (1 - self.influence) * self.filteredY[i - 1]
            self.avgFilter[i] = np.mean(self.filteredY[(i - self.lag):i])
            self.stdFilter[i] = np.std(self.filteredY[(i - self.lag):i])
        else:
            self.signals[i] = 0
            self.filteredY[i] = self.y[i]
            self.avgFilter[i] = np.mean(self.filteredY[(i - self.lag):i])
            self.stdFilter[i] = np.std(self.filteredY[(i - self.lag):i])

        return self.signals[i]

以下是平滑z-score算法的Scala版本(非惯用):

/**
  * Smoothed zero-score alogrithm shamelessly copied from https://stackoverflow.com/a/22640362/6029703
  * Uses a rolling mean and a rolling deviation (separate) to identify peaks in a vector
  *
  * @param y - The input vector to analyze
  * @param lag - The lag of the moving window (i.e. how big the window is)
  * @param threshold - The z-score at which the algorithm signals (i.e. how many standard deviations away from the moving mean a peak (or signal) is)
  * @param influence - The influence (between 0 and 1) of new signals on the mean and standard deviation (how much a peak (or signal) should affect other values near it)
  * @return - The calculated averages (avgFilter) and deviations (stdFilter), and the signals (signals)
  */
private def smoothedZScore(y: Seq[Double], lag: Int, threshold: Double, influence: Double): Seq[Int] = {
  val stats = new SummaryStatistics()

  // the results (peaks, 1 or -1) of our algorithm
  val signals = mutable.ArrayBuffer.fill(y.length)(0)

  // filter out the signals (peaks) from our original list (using influence arg)
  val filteredY = y.to[mutable.ArrayBuffer]

  // the current average of the rolling window
  val avgFilter = mutable.ArrayBuffer.fill(y.length)(0d)

  // the current standard deviation of the rolling window
  val stdFilter = mutable.ArrayBuffer.fill(y.length)(0d)

  // init avgFilter and stdFilter
  y.take(lag).foreach(s => stats.addValue(s))

  avgFilter(lag - 1) = stats.getMean
  stdFilter(lag - 1) = Math.sqrt(stats.getPopulationVariance) // getStandardDeviation() uses sample variance (not what we want)

  // loop input starting at end of rolling window
  y.zipWithIndex.slice(lag, y.length - 1).foreach {
    case (s: Double, i: Int) =>
      // if the distance between the current value and average is enough standard deviations (threshold) away
      if (Math.abs(s - avgFilter(i - 1)) > threshold * stdFilter(i - 1)) {
        // this is a signal (i.e. peak), determine if it is a positive or negative signal
        signals(i) = if (s > avgFilter(i - 1)) 1 else -1
        // filter this signal out using influence
        filteredY(i) = (influence * s) + ((1 - influence) * filteredY(i - 1))
      } else {
        // ensure this signal remains a zero
        signals(i) = 0
        // ensure this value is not filtered
        filteredY(i) = s
      }

      // update rolling average and deviation
      stats.clear()
      filteredY.slice(i - lag, i).foreach(s => stats.addValue(s))
      avgFilter(i) = stats.getMean
      stdFilter(i) = Math.sqrt(stats.getPopulationVariance) // getStandardDeviation() uses sample variance (not what we want)
  }

  println(y.length)
  println(signals.length)
  println(signals)

  signals.zipWithIndex.foreach {
    case(x: Int, idx: Int) =>
      if (x == 1) {
        println(idx + " " + y(idx))
      }
  }

  val data =
    y.zipWithIndex.map { case (s: Double, i: Int) => Map("x" -> i, "y" -> s, "name" -> "y", "row" -> "data") } ++
    avgFilter.zipWithIndex.map { case (s: Double, i: Int) => Map("x" -> i, "y" -> s, "name" -> "avgFilter", "row" -> "data") } ++
    avgFilter.zipWithIndex.map { case (s: Double, i: Int) => Map("x" -> i, "y" -> (s - threshold * stdFilter(i)), "name" -> "lower", "row" -> "data") } ++
    avgFilter.zipWithIndex.map { case (s: Double, i: Int) => Map("x" -> i, "y" -> (s + threshold * stdFilter(i)), "name" -> "upper", "row" -> "data") } ++
    signals.zipWithIndex.map { case (s: Int, i: Int) => Map("x" -> i, "y" -> s, "name" -> "signal", "row" -> "signal") }

  Vegas("Smoothed Z")
    .withData(data)
    .mark(Line)
    .encodeX("x", Quant)
    .encodeY("y", Quant)
    .encodeColor(
      field="name",
      dataType=Nominal
    )
    .encodeRow("row", Ordinal)
    .show

  return signals
}

下面是一个测试，返回与Python和Groovy版本相同的结果:

val y = List(1d, 1d, 1.1d, 1d, 0.9d, 1d, 1d, 1.1d, 1d, 0.9d, 1d, 1.1d, 1d, 1d, 0.9d, 1d, 1d, 1.1d, 1d, 1d,
  1d, 1d, 1.1d, 0.9d, 1d, 1.1d, 1d, 1d, 0.9d, 1d, 1.1d, 1d, 1d, 1.1d, 1d, 0.8d, 0.9d, 1d, 1.2d, 0.9d, 1d,
  1d, 1.1d, 1.2d, 1d, 1.5d, 1d, 3d, 2d, 5d, 3d, 2d, 1d, 1d, 1d, 0.9d, 1d,
  1d, 3d, 2.6d, 4d, 3d, 3.2d, 2d, 1d, 1d, 0.8d, 4d, 4d, 2d, 2.5d, 1d, 1d, 1d)

val lag = 30
val threshold = 5d
val influence = 0d

smoothedZScore(y, lag, threshold, influence)

这里的要点

如果边界值或其他标准取决于未来值，那么唯一的解决方案(没有时间机器，或其他关于未来值的知识)是推迟任何决定，直到有足够的未来值。如果你想要一个高于均值的水平，例如，20点，那么你必须等到你至少有19点才能做出任何峰值决策，否则下一个新点可能会完全超过你19点之前的阈值。

Added: If the statistical distribution of the peak heights could be heavy tailed, instead of Uniform or Gaussian, then you may need to wait until you see several thousand peaks before it starts to become unlikely that a hidden Pareto distribution won't produce a peak many times larger than any you currently have seen before or have in your current plot. Unless you somehow know in advance that the very next point can't be 1e20, it could appear, which after rescaling your plot's Y dimension, would be flat up until that point.