python 平均值 预测_如何使用移动平均值/滚动平均值预测下一个值

我不认为滚动平均数对你有用，因为它不会向你显示任何趋势。假设你的股票每年上涨10美元，你的滚动平均数会严重低估你明年的股票价值。在

我建议使用线性外推法(例如最后使用的3个单位)import pylab

from numpy import polyfit, poly1d, linspace

import matplotlib.pyplot as plt

data = [[718394219, 2013 , 01],

[763723622, 2014 , 01],

[762225057, 2015 , 01],

[787479774, 2016 , 01],

[845614054, 2017 , 01]]

values = [i[0] for i in data]

years = [i[1] for i in data]

plt.plot(years, values)

for i in range(0, len(data)-3):

# interpolate values

# you can choose for yourself how many point should be taken into account

# as well as the degree of interpolation (the last variable mentioned)

z = polyfit(years[i:i+3],values[i:i+3], 1)

p = poly1d(z)

# extrapolate interpolation by 1

start_year = years[i]

end_year = years[i+3] + 1

# plot values

x = linspace(start_year, end_year, 10)

y = p(x)

plt.plot(x, y)

plt.show()

您可以更改此位以满足您的需要：

^{pr2}$

说得很清楚，我没有统计学的背景，如果“准确度不需要很高”，我就是这样预测的。在