zoukankan      html  css  js  c++  java
  • Exponentially Weighted Moving-Average

    指数加权平均值 一下是基于pandas的。

    应该是和exponential smoothing or Holt–Winters method 是基于一个基础的。

    DEMO, 原址:http://connor-johnson.com/2014/02/01/smoothing-with-exponentially-weighted-moving-averages/

    # -*- coding: utf-8 -*-
    """
    Created on Sat Nov 01 13:02:50 2014
    
    @author: dell
    """
    
    import pandas, numpy as np
    ewma = pandas.stats.moments.ewma
    import matplotlib.pyplot as plt
    
    # make a hat function, and add noise
    x = np.linspace(0,1,100)
    x = np.hstack((x,x[::-1]))
    x += np.random.normal( loc=0, scale=0.1, size=200 )
    plt.plot( x, alpha=0.4, label='Raw' )
    
    # take EWMA in both directions with a smaller span term
    fwd = ewma( x, span=15 ) # take EWMA in fwd direction
    bwd = ewma( x[::-1], span=15 ) # take EWMA in bwd direction
    c = np.vstack(( fwd, bwd[::-1] )) # lump fwd and bwd together
    c = np.mean( c, axis=0 ) # average
    
    # regular EWMA, with bias against trend
    plt.plot( ewma( x, span=20 ), 'b', label='EWMA, span=20' )
    
    # "corrected" (?) EWMA
    plt.plot( c, 'r', label='Reversed-Recombined' )
    
    plt.legend(loc=8)
    plt.savefig( 'ewma_correction.png', fmt='png', dpi=100 )

    DEMO II

    # -*- coding: utf-8 -*-
    """
    Created on Sat Nov 01 14:10:05 2014
    
    @author: dell
    """
    
    import numpy as np
    import matplotlib.pyplot as plt
     
    def holt_winters_second_order_ewma( x, span, beta ):
        N = x.size
        alpha = 2.0 / ( 1 + span )
        s = np.zeros(( N, ))
        b = np.zeros(( N, ))
        s[0] = x[0]
        for i in range( 1, N ):
            s[i] = alpha * x[i] + ( 1 - alpha )*( s[i-1] + b[i-1] )
            b[i] = beta * ( s[i] - s[i-1] ) + ( 1 - beta ) * b[i-1]
        return s
     
    # make a hat function, and add noise
    x = np.linspace(0,1,100)
    x = np.hstack((x,x[::-1]))
    x += np.random.normal( loc=0, scale=0.1, size=200 ) + 3.0
    plt.plot( x, alpha=0.4, label='Raw' )
     
    # holt winters second order ewma
    plt.plot( holt_winters_second_order_ewma( x, 10, 0.3 ), 'b', label='Holt-Winters' )
     
    plt.title('Holt-Winters' )
    plt.legend( loc=8 )
     
    plt.savefig( 'holt_winters.png', fmt='png', dpi=100 )
  • 相关阅读:
    poj 2773 利用欧拉函数求互质数
    poj3358:欧拉定理
    poj:2992 因子数量
    poj3696:同余方程,欧拉定理
    USACO5.4-Character Recognition
    hdu5017:补题系列之西安网络赛1011
    hdu5014:number sequence对称思想
    欧拉函数,欧拉定理例题整理
    POJ 3463 Sightseeing (次短路)
    POJ
  • 原文地址:https://www.cnblogs.com/hluo/p/4067149.html
Copyright © 2011-2022 走看看