豁蒙楼

python股票分析

发表于 更新于 分类于 Disqus：

【Python学习】股票分析系列（中文自制） by 大力牛肉粉

东方财富网资金流向

Convert curl syntax to Python, Ansible URI, MATLAB, Node.js, R, PHP, Strest, Go, Dart, JSON, Elixir, Rust

1
2
3
4
5
6
7
8
9
%matplotlib inline

import datetime as dt
import pandas as pd
import numpy as np
import pandas_datareader as pdr # 获取在线数据
import matplotlib.pyplot as plt
from matplotlib import style # style

获取数据

1
2
3
4
5
6
7
start = dt.datetime(2019, 8, 30)
end = dt.datetime(2020, 8, 30)

df = pdr.get_data_yahoo("TSLA", start, end)
df.to_csv("tsla.csv")

df = pd.read_csv("tsla.csv", parse_dates=True, index_col=0)
1
2
3
style.use("ggplot")

df.plot(figsize=(16,10),fontsize=16)
<AxesSubplot:xlabel='Date'>
svg

十日均线

1
2
3
df = pd.read_csv("tsla.csv", parse_dates=True, index_col=0)
df["10ma"] = df["Adj Close"].rolling(window=10, min_periods=0).mean()
df.head()
High Low Open Close Volume Adj Close 10ma
Date
2019-08-30 46.487999 44.841999 45.830002 45.122002 46603000.0 45.122002 45.122002
2019-09-03 45.790001 44.632000 44.816002 45.001999 26770500.0 45.001999 45.062000
2019-09-04 45.692001 43.841999 45.377998 44.136002 28805000.0 44.136002 44.753334
2019-09-05 45.959999 44.169998 44.500000 45.916000 36976500.0 45.916000 45.044001
2019-09-06 45.928001 45.034000 45.439999 45.490002 20947000.0 45.490002 45.133201

ax和subplot

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
plt.rcParams['font.sans-serif'] = ['SimHei']
# SimHei：微软雅黑
# FangSong：仿宋
fig, ax = plt.subplots(figsize=(10,6))
ax.plot(df.index, df["Adj Close"])
ax.plot(df.index, df["10ma"])
ax.legend(["Adj Close", "10ma"])
ax.set_title("Adj_Close & 10ma", fontsize=18)
ax.set_xlabel("Date", fontsize=18, fontfamily = 'DejaVu Sans', fontstyle="italic")
ax.set_ylabel("$", fontsize = "x-large", fontstyle="oblique")



# ax.minorticks_on()
ax.grid(which="minor", axis="both")
# ax.margins(0)
plt.show()

svg 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
fig, axs = plt.subplots(2, figsize=(10,6))
ax = axs[0]
ax2 = axs[1]

ax.plot(df.index, df["Adj Close"])
ax.plot(df.index, df["10ma"])
ax.legend(["Adj Close", "10ma"])
ax.set_title("Adj_Close & 10ma")
# ax.set_xlabel("Date",  fontfamily = 'DejaVu Sans', fontstyle="italic")
ax.set_ylabel("$", fontstyle="oblique")

ax2.bar(df.index, df["Volume"])
ax2.set_ylabel("Volume", fontstyle="oblique")
style.use("ggplot")
plt.show()


svg

蜡烛图

1
2
import matplotlib.dates as mdates
from mpl_finance import candlestick_ohlc #蜡烛图
1
2
3
4
5
df = pd.read_csv("tsla.csv", parse_dates=True, index_col=0)
df_ohlc = df["Adj Close"].resample("10D").ohlc() #open high low close
df_volume = df["Volume"].resample("10D").sum()
df_ohlc.reset_index(inplace=True)
df_ohlc.Date = df_ohlc.Date.map(mdates.date2num)
1
2
3
4
5
6
7
8
9
10
11
12
fig = plt.figure(figsize=(16,10))
ax1 = plt.subplot2grid((6,1), (0,0), rowspan=5, colspan=1)
ax2 = plt.subplot2grid((6,1), (5,0), rowspan=1, colspan=1, sharex = ax1)

ax1.xaxis_date()
candlestick_ohlc(ax1, df_ohlc.values, width=5, colorup='g')
ax2.fill_between(df_volume.index.map(mdates.date2num), df_volume.values, 0)
ax2.plot(df_volume)

style.use("ggplot")

plt.show()
svg

标普500公司简称获取(爬虫)

1
2
3
from bs4 import BeautifulSoup as bs # 字典
import pickle # 列表，序列化
import requests # 网络请求
1
2
3
4
5
6
7
8
9
10
11
12
13
14
def save_sp500_tickers():
    resp = requests.get("https://en.wikipedia.org/wiki/List_of_S%26P_500_companies")
    soup = bs(resp.text, "lxml")
    table = soup.find("table", {"class":"wikitable sortable"})
    tickers = []
    for row in table.findAll('tr')[1:]:
        ticker = row.findAll('td')[0].text.replace("\n", "")
        tickers.append(ticker)
    # print(tickers)
    with open("sp500tickers.pickle","wb") as f: # 二进制只读
        pickle.dump(tickers, f)
    return tickers
        
# save_sp500_tickers()

下载SP500股票数据

1
2
3
4
5
import os # 系统接口
import pandas as pd 
import pandas_datareader as pdr
import datetime as dt

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
def get_data_from_yahoo(isReload=False):
    if isReload:
        tickers = save_sp500_tickers()
    else:
        with open("sp500tickers.pickle","rb") as f:
            tickers = pickle.load(f)
    
    if not os.path.exists("stock_dfs"): #当前目录无该子目录
        os.makedirs("stock_dfs")
    
    start = dt.datetime(2016, 1, 1)
    end = dt.datetime(2019, 1, 1)

    for ticker in tickers:
        if not os.path.exists("stock_dfs/{}.csv".format(ticker)):
            try:
                df = pdr.get_data_yahoo(ticker, start, end)
                df.to_csv("stock_dfs/{}.csv".format(ticker))
                print("Download complete: {}".format(ticker))
            except:
                continue
        else:
            print("Already exist: {}".format(ticker))
    
get_data_from_yahoo(True)

整合数据

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
def compile_data():
    with open("sp500tickers.pickle",'rb') as f:
        tickers = pickle.load(f)[:10]

    main_df = pd.DataFrame()

    # compile
    for count, ticker in enumerate(tickers):
        df = pd.read_csv("stock_dfs/{}.csv".format(ticker))
        df.set_index("Date", inplace=True)
        df.rename(columns = {"Adj Close": ticker}, inplace=True)
        df.drop(["Open", "High", "Low", "Close", "Volume"], axis=1, inplace=True)
        
        if main_df.empty:
            main_df = df
        else:
            main_df = main_df.join(df, how="left")
        print(count)
        if count == 10:
            break
    main_df.to_csv("sp500_joined.csv")
    return main_df

main_df = compile_data()
0
1
2
3
4
5
6
7
8
9
1
main_df
MMM ABT ABBV ABMD ACN ATVI ADBE AMD AAP AES
Date
2016-01-04 128.033249 39.050327 46.141216 85.239998 93.774796 36.365803 91.970001 2.770000 150.250793 7.769172
2016-01-05 128.591339 39.041222 45.948997 85.000000 94.262863 35.901806 92.339996 2.750000 149.224380 7.876278
2016-01-06 126.001358 38.713760 45.956993 85.300003 94.078682 35.563477 91.019997 2.510000 145.276642 7.604397
2016-01-07 122.931786 37.785950 45.820843 81.919998 91.316017 35.060810 89.110001 2.280000 146.885345 7.414905
2016-01-08 122.513206 36.994564 44.571400 84.580002 90.431938 34.519478 87.849998 2.140000 143.658066 7.522010
... ... ... ... ... ... ... ... ... ... ...
2018-12-24 168.035690 63.701054 75.395958 281.079987 130.427887 43.354313 205.160004 16.650000 147.761047 13.031178
2018-12-26 175.222961 67.645943 79.767776 307.440002 135.638382 45.749195 222.949997 17.900000 153.823486 13.464924
2018-12-27 179.399902 68.627304 80.547180 315.670013 137.004425 46.360237 225.139999 17.490000 153.486160 13.474353
2018-12-28 178.148697 69.074249 81.631172 318.170013 136.428741 46.123703 223.130005 17.820000 154.250168 13.455495
2018-12-31 179.249359 70.279106 82.589752 325.040009 137.589874 45.897026 226.240005 18.459999 156.234573 13.634648

754 rows × 10 columns

数据可视化

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
# 股票相关度？
def visualize_data():
    df = pd.read_csv("sp500_joined.csv")
    df_corr = df.corr()
    df_corr.to_csv("sp500corr.csv")
    data1 = df_corr.values
    fig1 = plt.figure()
    ax1 = fig1.add_subplot(111)

    heatmap1 = ax1.pcolor(data1, cmap=plt.cm.RdYlGn)
    fig1.colorbar(heatmap1)

    ax1.set_xticks(np.arange(data1.shape[1]))
    ax1.set_yticks(np.arange(data1.shape[0]))
    ax1.invert_yaxis()
    ax1.xaxis.tick_top()
    column_labels = df_corr.columns
    row_labels = df_corr.index
    ax1.set_xticklabels(column_labels)
    ax1.set_yticklabels(row_labels)
    plt.xticks(rotation=90)
    heatmap1.set_clim(-1,1)


visualize_data()

svg 
1
import seaborn as sns
1
2
3
4
5
6
7
8
9
10
11
12
13
# 股票相关度？
def visualize_data():
    df = pd.read_csv("sp500_joined.csv")
    df_corr = df.corr()
    df_corr.to_csv("sp500corr.csv")
    
    f, axes = plt.subplots(1, figsize=(10,8))
    # sns.set(style="whitegrid")
    sns.despine(left=True)
    sns.heatmap(df_corr, cmap= "RdYlGn", cbar=True)

visualize_data()

svg

获取7日内股价波动

  • 买/卖/持有？
  • 7日内表现
1
2
3
4
5
6
7
8
9
def process_data_for_labels(ticker):
    hm_days = 7
    df = pd.read_csv("sp500_joined.csv", index_col=0)
    tickers = df.columns.values.tolist()
    df.fillna(0, inplace=True)
    for i in range(1, hm_days+1):
        df["{}_{}d".format(ticker, i)] = (df[ticker].shift(-i)-df[ticker])/df[ticker]
    df.fillna(0, inplace=True)
    return tickers, df
1
2
3
4
5
6
7
8
9
10
11
def buy_sell_hold(*args):
    cols = [c for c in args]
    requirement = 0.02
    for col in cols:
        if col > requirement:
            return 1
        elif col < -requirement:
            return -1
        else:
            return 0

获取特征集合

1
from collections import Counter
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
def extract_featuresets(ticker):
    tickers, df = process_data_for_labels(ticker)
    df["{}_target".format(ticker)] = list(map(buy_sell_hold, 
                                    df["{}_1d".format(ticker)],
                                    df["{}_2d".format(ticker)],
                                    df["{}_3d".format(ticker)],
                                    df["{}_4d".format(ticker)],
                                    df["{}_5d".format(ticker)],
                                    df["{}_6d".format(ticker)],
                                    df["{}_7d".format(ticker)],
                                    ))

    vals = df["{}_target".format(ticker)].values.tolist()
    str_vals = [str(i) for i in vals]
    # print("Data spread", Counter(str_vals))

    df.fillna(0, inplace=True)
    df.replace([np.inf, -np.inf], np.nan, inplace=True)
    df.dropna(inplace=True)

    df_vals = df[[t for t in tickers]].pct_change() # 环比
    df_vals.replace([np.inf, -np.inf], np.nan)
    df_vals.fillna(0, inplace=True)

    X = df_vals.values
    y = df[f"{ticker}_target"].values

    return X, y, df

机器学习

1
X, y, _ = extract_featuresets("MMM")
1
2
3
from sklearn.model_selection import train_test_split
from sklearn import svm, neighbors #cross_validation deprecated 
from sklearn.ensemble import VotingClassifier, RandomForestClassifier# 综合不同算法
1
2
def do_ml(ticker):
    X,y,df