使用Python分析股价波动周期

基本思路是获取股价收盘信息后，使用希尔伯特黄变换将股价波动数据拆解为不同周期的波动曲线。再本别利用频谱分析计算每一个曲线的频率。目标是将股价波动数据拆解为不同周期波动的叠加态。

1.获取收盘价

富途有很好的API接口，给我这种小散送了每个月的使用次数也够了。

富途openAPI官网

2.希尔伯特黄变换

利用pyhht包，官方的文档磕磕绊绊看懂。

 

合起来

import pyhht
from pyhht.visualization import plot_imfs
import numpy as np
import random
from futu import *
import pandas as pd
import sys
import re
def likaiHHT_savefig_imfs(filepath,xlabel,title,signal, imfs, time_samples=None, fignum=None):
    if time_samples is None:
        time_samples = np.arange(signal.shape[0])

    n_imfs = imfs.shape[0]

    plt.figure(num=fignum)
    axis_extent = max(np.max(np.abs(imfs[:-1, :]), axis=0))

    # Plot original signal
    ax = plt.subplot(n_imfs + 1, 1, 1)
    ax.plot(time_samples, signal)
    ax.axis([time_samples[0], time_samples[-1], signal.min(), signal.max()])
    ax.tick_params(which='both', left=True, bottom=False, labelleft=True,labelbottom=False)
    ax.grid(False)
    ax.set_ylabel('signal')
    ax.set_title(title)

    # Plot the IMFs
    for i in range(n_imfs - 1):
        print(i + 2)
        ax = plt.subplot(n_imfs + 1, 1, i + 2)
        ax.plot(time_samples, imfs[i, :])
        ax.axis([time_samples[0], time_samples[-1], -axis_extent, axis_extent])
        ax.tick_params(which='both', left=True, bottom=False, labelleft=True,labelbottom=False)
        ax.grid(False)
        ax.set_ylabel('imf' + str(i + 1))

    # Plot the residue
    ax = plt.subplot(n_imfs + 1, 1, n_imfs + 1)
    ax.plot(time_samples, imfs[-1, :], 'r')
    ax.axis('auto')
    #ax.tick_params(which='both', left=False, bottom=False, labelleft=False,labelbottom=False)
    ax.grid(False)
    ax.set_ylabel('res.')
    ax.set_xlabel(xlabel)
    plt.savefig(filepath)
    return 
def imfs_max_freq(imfs,sample_rate,fft_size):
#计算每一个imfs频谱中最高的那个频率
    n_imfs=imfs.shape[0]
    max_freq=[]
    for i in range(n_imfs-1):
        xs=imfs[i,:][:fft_size]
        xf=np.fft.rfft(xs)/fft_size
        freqs=np.linspace(0,sample_rate/2,fft_size//2+1)
        xfp=20*np.log10(np.clip(np.abs(xf), 1e-20, 1e100))
        max_freq.append(freqs[np.argmax(xfp)])
    return max_freq
def HHTstock(stockid,begindate,enddate):
    closelist=[]
    quote_ctx = OpenQuoteContext(host='127.0.0.1', port=11111)  # 创建行情对象
    ret, data, page_req_key = quote_ctx.request_history_kline(stockid, start=begindate, end=enddate, max_count=5)  # 每页5个，请求第一页
    if ret == RET_OK:
        #print(data)
        #print(data['code'][0])    # 取第一条的股票代码
        #print(data['close'].values.tolist())   # 第一页收盘价转为list
        closelist=data['close'].values.tolist()
    else:
        print('error:', data)
    while page_req_key != None:  # 请求后面的所有结果
        #print('*************************************')
        ret, data, page_req_key = quote_ctx.request_history_kline(stockid, start=begindate, end=enddate, max_count=5, page_req_key=page_req_key) # 请求翻页后的数据
        if ret == RET_OK:
            #print(data)
            closelist.extend(data['close'].values.tolist())
        else:
            print('error:', data)
    print('All pages are finished!')
    quote_ctx.close() # 关闭对象，防止连接条数用尽
    trading_day_num=len(closelist)
    t=np.linspace(0,trading_day_num,trading_day_num)
    np_close=np.array(closelist)
    decomposer=pyhht.EMD(np_close)
    imfs=decomposer.decompose()
    #plot_imfs(np_close,imfs,t)
    likaiHHT_savefig_imfs('./'+stockid+'.png','t/day',stockid,np_close,imfs,t)
    ls=imfs_max_freq(imfs,1,1000)#算每一段曲线的频率
    print(ls)
    return;
def main():
    HHTstock('HK.01816','2012-9-11','2020-9-18')
if __name__ == '__main__':
    main()

计算结果