平安股票分析

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
%matplotlib inline

import tushare as ts

# 601318 中国平安
pingan = ts.get_k_data('601318',start='2011-01-01')

pingan.head()

	date	open	close	high	low	volume	code
0	2011-01-04	24.984	25.011	25.314	24.830	245626.0	601318
1	2011-01-05	24.870	24.110	24.962	24.083	427554.0	601318
2	2011-01-06	24.145	23.112	24.145	22.611	947078.0	601318
3	2011-01-07	23.279	23.297	23.842	22.765	659013.0	601318
4	2011-01-10	23.248	22.866	23.688	22.809	282919.0	601318

pingan.tail()

	date	open	close	high	low	volume	code
2018	2019-05-06	82.80	81.20	83.60	79.66	1808609.0	601318
2019	2019-05-07	81.78	81.17	82.28	79.89	974951.0	601318
2020	2019-05-08	78.73	78.78	80.55	78.38	868375.0	601318
2021	2019-05-09	78.26	76.66	78.26	76.11	1125721.0	601318
2022	2019-05-10	77.95	81.38	81.87	77.30	1392098.0	601318

pingan.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 2023 entries, 0 to 2022
Data columns (total 7 columns):
date      2023 non-null object
open      2023 non-null float64
close     2023 non-null float64
high      2023 non-null float64
low       2023 non-null float64
volume    2023 non-null float64
code      2023 non-null object
dtypes: float64(5), object(2)
memory usage: 126.4+ KB

# 将data变成index
pingan['date'] = pd.to_datetime(pingan.date)

pingan.set_index('date',inplace = True)

pingan.index

DatetimeIndex(['2011-01-04', '2011-01-05', '2011-01-06', '2011-01-07',
               '2011-01-10', '2011-01-11', '2011-01-12', '2011-01-13',
               '2011-01-14', '2011-01-17',
               ...
               '2019-04-24', '2019-04-25', '2019-04-26', '2019-04-29',
               '2019-04-30', '2019-05-06', '2019-05-07', '2019-05-08',
               '2019-05-09', '2019-05-10'],
              dtype='datetime64[ns]', name='date', length=2023, freq=None)

pingan.head()

	open	close	high	low	volume	code
date
2011-01-04	24.984	25.011	25.314	24.830	245626.0	601318
2011-01-05	24.870	24.110	24.962	24.083	427554.0	601318
2011-01-06	24.145	23.112	24.145	22.611	947078.0	601318
2011-01-07	23.279	23.297	23.842	22.765	659013.0	601318
2011-01-10	23.248	22.866	23.688	22.809	282919.0	601318

pingan['close'].plot(figsize=(12,8))

<matplotlib.axes._subplots.AxesSubplot at 0x1a3adcc3c50>

# 重新以月采样 这个ohlc对应的是股市中的open，high，low，close这几个价格。
g_yue = pingan['close'].resample('M').ohlc()
g_yue.head()

	open	high	low	close
date
2011-01-31	25.011	25.011	21.249	21.842
2011-02-28	21.952	22.980	21.477	21.868
2011-03-31	22.084	23.512	21.433	21.736
2011-04-30	22.514	24.123	22.514	22.954
2011-05-31	23.103	23.103	20.972	21.407

pingan['close'].loc['2017'].plot(figsize=(12,8))

<matplotlib.axes._subplots.AxesSubplot at 0x1a3ae7719b0>

# 计算总收益率
shouyi = (pingan['close'][-1] - pingan['close'][0])/pingan['close'][0]
shouyi

2.253768341929551

# 年数
y_num = pingan.index[-1].year - pingan.index[0].year
y_num

8

# 年收益率
shouyi**(1/y_num)

1.1069134365512923

# 画一年为单位的 close
pingan['close'].to_period('A').groupby(level = 0).first().plot()

<matplotlib.axes._subplots.AxesSubplot at 0x1a3ae771748>

# 计算移动窗口42 和250 的均值
pingan['42d'] = pingan['close'].rolling(window=42).mean()
pingan['250d'] = pingan['close'].rolling(window=250).mean()

pingan[['close','42d','250d']].plot(figsize=(12,8),color = ['y','b','r'])

<matplotlib.axes._subplots.AxesSubplot at 0x1a3ae784c88>

# 对数收益率
pingan['duishu'] = np.log(pingan['close']/pingan['close'].shift(1))   #shifting 指的是沿着时间轴将数据前移或后移。

pingan.head()

	open	close	high	low	volume	code	42d	250d	duishu
date
2011-01-04	24.984	25.011	25.314	24.830	245626.0	601318	NaN	NaN	NaN
2011-01-05	24.870	24.110	24.962	24.083	427554.0	601318	NaN	NaN	-0.036689
2011-01-06	24.145	23.112	24.145	22.611	947078.0	601318	NaN	NaN	-0.042275
2011-01-07	23.279	23.297	23.842	22.765	659013.0	601318	NaN	NaN	0.007973
2011-01-10	23.248	22.866	23.688	22.809	282919.0	601318	NaN	NaN	-0.018674

pingan['duishu'].plot(subplots = True,figsize=(12,12))

array([<matplotlib.axes._subplots.AxesSubplot object at 0x000001A3AEDBE4E0>],
      dtype=object)

波动挺大的