# ** 5-4 Finance Analysis** Technical Indicator | https://wikidocs.net/3396

### **01 Pandas_datareader** > **from** pandas_datareader **import** get_data_yahoo

In [1]:

# from pandas_datareader import get_data_yahoo
# price = get_data_yahoo('005930.KS','2017-06-01')
# price = price.fillna('ffill')
# price = price.dropna()
# price.tail(3)

In [2]:

# 저장된 CSV 파일을 사용한다

import pandas as pd
price = pd.read_csv('./data/stock.csv')
price = price.set_index('Date')
price.head(3)
price.index = pd.DatetimeIndex(price.index)
price.info()

<class 'pandas.core.frame.DataFrame'>
DatetimeIndex: 248 entries, 2017-06-01 to 2018-06-11
Data columns (total 6 columns):
High         248 non-null float64
Low          248 non-null float64
Open         248 non-null float64
Close        248 non-null float64
Volume       248 non-null float64
Adj Close    248 non-null float64
dtypes: float64(6)
memory usage: 13.6 KB

## **02 Stochastic Oscillator** 주어진 기간중 당일종가가 상대적 위치를 판단하는 지표

In [3]:

def fnStoch(price, n=14):
    import pandas as pd
    date_num = len(price['Close'])
    tempSto_K = []
    for i in range(date_num):
        if i >= n -1:
            tempUp   = price['Close'][i] - min(price['Low'][i+1 -n : i+1])
            tempDown = max(price['High'][i+1 -n : i+1]) -  min(price['Low'][i+1 -n : i+1])
            tempSto_K.append( tempUp / tempDown )
        else:
            tempSto_K.append(0) 
    price['Sto_K']     = pd.Series(tempSto_K ,   index=price.index)
    price['Sto_D']     = pd.Series(price['Sto_K'].rolling(3).mean())
    price['Sto_SlowD'] = pd.Series(price['Sto_D'].rolling(3).mean())
    return price

In [4]:

price = fnStoch(price)
price.tail(3)

Out[4]:

	High	Low	Open	Close	Volume	Adj Close	Sto_K	Sto_D	Sto_SlowD
Date
2018-06-07	51800.0	50500.0	51800.0	50600.0	13590016.0	50600.0	0.384615	0.487179	0.509972
2018-06-08	50400.0	49600.0	50200.0	49650.0	16951706.0	49650.0	0.141026	0.363248	0.465812
2018-06-11	50300.0	49350.0	49750.0	49900.0	10496083.0	49900.0	0.205128	0.243590	0.364672

In [5]:

%matplotlib inline
import matplotlib.pyplot as plt
plt.figure(figsize=(25,4))

price.Sto_D.plot()
price.Sto_K.plot()
plt.axhline(y = 0.25, linestyle = '--', color = 'red')
plt.axhline(y = 0.75, linestyle = '--', color = 'blue')

Out[5]:

<matplotlib.lines.Line2D at 0x7fd9a3420d68>

## **3 RSI(Relative Strength Index; 상대강도지수)** 일정기간 가격상승압력과 하락압력간의 상대적인 강도를 판단한다

In [6]:

# np.diff()
import pandas as pd
data = [i   for i in range(100)]
data = pd.Series(data)
data[:10]

Out[6]:

0    0
1    1
2    2
3    3
4    4
5    5
6    6
7    7
8    8
9    9
dtype: int64

In [7]:

data.diff(4)[:10]

Out[7]:

0    NaN
1    NaN
2    NaN
3    NaN
4    4.0
5    4.0
6    4.0
7    4.0
8    4.0
9    4.0
dtype: float64

In [8]:

def fnRSI(price, m_N):    
    import numpy as np
    import pandas as pd

    m_Df = price.Close
    U = np.where(m_Df.diff(1) > 0, m_Df.diff(1), 0)
    D = np.where(m_Df.diff(1) < 0, m_Df.diff(1) *(-1), 0)

    AU = pd.Series(U).rolling( window=m_N).mean()
    AD = pd.Series(D).rolling( window=m_N).mean()
    RSI = AU.div(AD + AU) * 100
    RSI.index = price.index
    price['RSI'] = RSI.fillna(0)
    return price

In [9]:

price = fnRSI(price, 3)
price.tail(3)

Out[9]:

	High	Low	Open	Close	Volume	Adj Close	Sto_K	Sto_D	Sto_SlowD	RSI
Date
2018-06-07	51800.0	50500.0	51800.0	50600.0	13590016.0	50600.0	0.384615	0.487179	0.509972	18.181818
2018-06-08	50400.0	49600.0	50200.0	49650.0	16951706.0	49650.0	0.141026	0.363248	0.465812	10.810811
2018-06-11	50300.0	49350.0	49750.0	49900.0	10496083.0	49900.0	0.205128	0.243590	0.364672	13.157895

In [10]:

price['RSI'].plot(figsize=(25,4), label='RSI', color='black')
plt.axhline(y = 30, linestyle = '--', color = 'blue', label = 'Under')
plt.axhline(y = 70, linestyle = '--', color = 'red',  label = 'Over') 
plt.legend()

Out[10]:

<matplotlib.legend.Legend at 0x7fd9a14f47b8>

## **4 볼린저 밴드 데이터 그리기** Bollinger Bands

In [11]:

def fnBolingerBand(price, n=20, k=2):
    price['20d_ma']    = price['Close'].rolling(n).mean()
    price['Bol_upper'] = price['Close'].rolling(n).mean() + \
                         k * price['Close'].rolling(n, min_periods=n).std()
    price['Bol_lower'] = price['Close'].rolling(n).mean() - \
                         k * price['Close'].rolling(n, min_periods=n).std()
    return price

In [12]:

price = fnBolingerBand(price)
price.tail(3)

Out[12]:

	High	Low	Open	Close	Volume	Adj Close	Sto_K	Sto_D	Sto_SlowD	RSI	20d_ma	Bol_upper	Bol_lower
Date
2018-06-07	51800.0	50500.0	51800.0	50600.0	13590016.0	50600.0	0.384615	0.487179	0.509972	18.181818	50792.5	52789.412090	48795.587910
2018-06-08	50400.0	49600.0	50200.0	49650.0	16951706.0	49650.0	0.141026	0.363248	0.465812	10.810811	50730.0	52789.994890	48670.005110
2018-06-11	50300.0	49350.0	49750.0	49900.0	10496083.0	49900.0	0.205128	0.243590	0.364672	13.157895	50645.0	52694.107636	48595.892364

In [13]:

plt.figure(figsize=(15,9))
price.Bol_upper.plot()
price.Bol_lower.plot()
price.Close.plot()
plt.legend()
plt.grid()
plt.show()

## **5 MACD(moving average convergence/divergence)** np.ewm()

In [14]:

def fnMACD(price, m_NumFast=12, m_NumSlow=26, m_NumSignal=9):
    price['EMAFast']    = price['Close'].rolling(m_NumFast).mean()
    price['EMASlow']    = price['Close'].rolling(m_NumSlow).mean()
    price['MACD']       = price['EMAFast'] - price['EMASlow']
    price['MACDSignal'] = price['MACD'].ewm( span = m_NumSignal, min_periods = m_NumSignal-1).mean()
    price['MACDDiff']   = price['MACD'] - price['MACDSignal']
    return price

In [15]:

price = fnMACD(price)
price.tail(3)

Out[15]:

	High	Low	Open	Close	Volume	Adj Close	Sto_K	Sto_D	Sto_SlowD	RSI	20d_ma	Bol_upper	Bol_lower	EMAFast	EMASlow	MACD	MACDSignal	MACDDiff
Date
2018-06-07	51800.0	50500.0	51800.0	50600.0	13590016.0	50600.0	0.384615	0.487179	0.509972	18.181818	50792.5	52789.412090	48795.587910	51166.666667	51244.230769	-77.564103	-312.104566	234.540463
2018-06-08	50400.0	49600.0	50200.0	49650.0	16951706.0	49650.0	0.141026	0.363248	0.465812	10.810811	50730.0	52789.994890	48670.005110	51137.500000	51115.384615	22.115385	-245.260576	267.375960
2018-06-11	50300.0	49350.0	49750.0	49900.0	10496083.0	49900.0	0.205128	0.243590	0.364672	13.157895	50645.0	52694.107636	48595.892364	50979.166667	50996.153846	-16.987179	-199.605897	182.618717

In [16]:

%matplotlib inline
plt.figure(figsize=(25,4))
price.MACD.plot()
price.MACDSignal.plot()
price.MACDDiff.plot()
plt.axhline(y = 0, linestyle = '--', color = 'k')
plt.legend()
plt.grid()
plt.show()