-
Notifications
You must be signed in to change notification settings - Fork 47
/
indicator.py
427 lines (340 loc) · 13.8 KB
/
indicator.py
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
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
# All formula : reference at http://stockcharts.com/school/doku.php?id=chart_school/
import pandas as pd
import numpy as np
def roc(df_close, periods=12):
# Close - Close n periods ago
# change_n_period_ago = df_close - df_close.shift(periods)
change_n_period_ago = df_close.diff(periods)
# Close n periods ago
close_n_period_ago = df_close.shift(periods)
#ROC = [(Close - Close n periods ago) / (Close n periods ago)] * 100
ROC = change_n_period_ago / close_n_period_ago * 100
return ROC
def is_strong(df_close, window=12):
""" In 12 days periods, How many are ROC'security more than ROC' SET (percent %) """
# df_roc has calculated by daily_returns
df_roc = roc(df_close)
# Empty Data Frame
df_main = pd.DataFrame(index=df_close.index, columns=df_close.columns)
for symbol in df_close.columns:
if symbol == 'SET':
continue
df_compare = df_roc['SET'] < df_roc[symbol]
# In python True is 1, False is 0
# 1: meaning 12 days periods, ROC'security > ROC'SET always
# 2: meaning 12 days periods, ROC'security < ROC'SET always
df_main[symbol] = pd.rolling_mean(df_compare[window:], window)
df_main['SET']=0
print(df_main)
def get_rolling_mean(values, window):
"""Return rolling mean of given values, using specified window size."""
#return pd.rolling_mean(values, window=window)
return values.rolling(center=False, window=window).mean()
def get_rolling_std(values, window):
"""Return rolling standard deviation of given values, using specified window size."""
# Compute and return rolling standard deviation
#return pd.rolling_std(values, window=window)
return values.rolling(center=False,window=window).std()
def get_bollinger_bands(rm, rstd):
"""Return upper and lower Bollinger Bands."""
# Compute upper_band and lower_band
upper_band = rm + rstd * 2
lower_band = rm - rstd * 2
return upper_band, lower_band
def daily_returns(df):
"""Compute and return the daily return values."""
# (current_price / previous_price) -1
# daily_returns = (df / df.shift(1)) - 1
daily_returns = df.pct_change();
daily_returns.iloc[0] = 0
return daily_returns
def close_2_open(df):
"""Compute and return the daily return values."""
# from pervious close to current open close
# (current_open_price / previous_close_price) -1
current_open = df['OPEN']
previos_close = df['CLOSE'].shift(1)
daily_returns_2 = (current_open/ previos_close) - 1
#daily_returns = df.pct_change();
#daily_returns_2.iloc[0] = 0
return pd.DataFrame(daily_returns_2, columns=['C2O'])
def BBANDS(df_price, periods=20, mul=2):
# Middle Band = 20-day simple moving average (SMA)
df_middle_band = pd.rolling_mean(df_price, window=periods)
#df_middle_band = pd.rolling(window=periods,center=False).mean()
# 20-day standard deviation of price
""" Pandas uses the unbiased estimator (N-1 in the denominator),
whereas Numpy by default does not.
To make them behave the same, pass ddof=1 to numpy.std()."""
df_std = pd.rolling_std(df_price, window=periods)
#df_std = pd.rolling(window=periods,center=False).std()
# Upper Band = 20-day SMA + (20-day standard deviation of price x 2)
df_upper_band = df_middle_band + (df_std * mul)
# Lower Band = 20-day SMA - (20-day standard deviation of price x 2)
df_lower_band = df_middle_band - (df_std * mul)
return (df_upper_band, df_middle_band, df_lower_band)
def get_BBANDS(df, periods=20, mul=2):
(upper, middle, lower) = BBANDS(df, periods, mul)
df_BBANDS = pd.concat([upper, middle, lower], axis=1, join='inner')
df_BBANDS.columns = ['UPPER', 'MIDDLE', 'LOWER']
return df_BBANDS
def get_myRatio(df_price, periods=20):
# Middle Band = 20-day simple moving average (SMA)
#df_middle_band = pd.rolling_mean(df_price, window=periods)
df_middle_band = df_price.rolling(center=False, window=periods).mean()
# 20-day standard deviation of price
""" Pandas uses the unbiased estimator (N-1 in the denominator),
whereas Numpy by default does not.
To make them behave the same, pass ddof=1 to numpy.std()."""
#df_std = pd.rolling_std(df_price, window=periods)
df_std = df_price.rolling(center=False, window=periods).std()
return (df_price - df_middle_band)/(df_std * 2)
def sma(df, periods=12):
# compute simple moving average
#return pd.rolling_mean(df, window=periods)
return df.rolling(center=False, window=periods).mean()
# not sure
def ema(df, periods=12):
# compute exponential moving average
#return pd.ewma(df, span = periods)
return df.ewm(span=periods, adjust=True, min_periods=0, ignore_na=False).mean()
def average_convergence(df, period_low=26, period_fast=12):
"""
compute the MACD (Moving Average Convergence/Divergence)
using a fast and slow exponential moving average'
"""
emaslow = ema(df, period_low)
emafast = ema(df, period_fast)
return (emaslow, emafast, emafast - emaslow)
def signal_MACD(df_MACD, periods=9):
return ema(df_MACD, periods)
def rsi(df):
periods=14
# Price change, df_change = df - df.shift(1)
df_change = df.diff(1)
df_gain = df_change.where(df_change > 0) # Gain
df_loss = -1 * df_change.where(df_change < 0) # loss, multiple -1 to positive values
df_gain.fillna(0, inplace=True) # fill NaN to 0
df_loss.fillna(0, inplace=True) # fill NaN to 0
# create DataFrame for saving Average Gain and Average Loss
df_avg_gain = pd.DataFrame(columns = df.columns, index = df.index)
df_avg_loss = df_avg_gain.copy()
df_avg_gain.iloc[periods] = df_gain[1:periods+1].mean() # First Average Gain = Sum of Gains over the past 14 periods / 14.
df_avg_loss.iloc[periods] = df_loss[1:periods+1].mean() # First Average Loss = Sum of Losses over the past 14 periods / 14
for index in range(periods+1, len(df)):
#Average Gain = [(previous Average Gain) x 13 + current Gain] / 14.
df_avg_gain.iloc[index] = (df_avg_gain.iloc[index-1] * 13 + df_gain.iloc[index])/periods
#Average Loss = [(previous Average Loss) x 13 + current Loss] / 14.
df_avg_loss.iloc[index] = (df_avg_loss.iloc[index-1] * 13 + df_loss.iloc[index])/periods
# RS = Average Gain / Average Loss
# if coding as bellow, it has a bug when df_avg_loss is zero (can't divid with zero)
# RS = df_avg_gain/df_avg_loss
# But I change coding with for loop instead
RS = pd.DataFrame(columns = df.columns, index = df.index)
for index in RS.index:
for sym in df.columns:
lossValue = df_avg_loss.loc[index][sym]
if lossValue == 0:
RS.loc[index][sym] = 100
else:
RS.loc[index][sym] = df_avg_gain.loc[index][sym]/lossValue
# 100
# RSI = 100 - --------
# 1 + RS
RSI = 100 - 100/(1 + RS)
return RSI
def sharpe_ratio(rp, rf=None):
if rf is None:
rf = rp.copy()
rf.iloc[0:] = 0
# rp = Expected porfolio return
# rf = Risk free rate
ret = rp - rf
# Sharpe ratio = mean(Expected porfolio return - Risk free rate)/Portfolio standard deviation
return ret.mean()/ret.std()
def rolling_sharpe_ratio(rp, rf, window):
# Example
# rp = df_daily_return[symbol]
# rf = df_daily_return['SET']
ret = rp - rf
mean = get_rolling_mean(ret, window)
std = get_rolling_std(ret, window )
# Sharpe ratio = mean(Expected porfolio return - Risk free rate)/Portfolio standard deviation
return mean/std
def create_dataframe_SR(df, symbols, window=5):
df_sr = pd.DataFrame(columns = symbols, index= df.index)
df_daily_return = daily_returns(df)
# Use SET for return reference, not use risk-free rate realy as return reference
rf = df_daily_return['SET']
for sym in symbols :
df_sr[sym] = rolling_sharpe_ratio(df_daily_return[sym], rf, window)
return df_sr
def true_range(df):
high = df['HIGH']
low = df['LOW']
previous_close = df['CLOSE'].shift(1)
# True Range (TR)
# Method 1: Current High less the current Low
# Method 2: Current High less the previous Close (absolute value)
# Method 3: Current Low less the previous Close (absolute value)
method1 = high-low
method2 = (high - previous_close).abs()
method3 = (low - previous_close).abs()
df_TR = pd.concat([method1, method2, method3], axis=1, join='inner')
#df_TR = pd.DataFrame(df_TR.max(axis=1), columns=['<TR>'])
return df_TR.max(axis='columns')
def ATR(df):
periods = 14
df_TR = true_range(df) # True range
df_ATR = pd.DataFrame(columns = ['ATR'], index = df.index)
df_ATR.iloc[periods-1] = df_TR[0:periods].mean() # First ATR = Sum of TR over the past 14 periods / 14
for index in range(periods, len(df)):
#Current ATR = [(Prior ATR x 13) + Current TR] / 14
# - Multiply the previous 14-day ATR by 13.
# - Add the most recent day's TR value.
# - Divide the total by 14
df_ATR.iloc[index] = (df_ATR.iloc[index-1] * 13 + df_TR.iloc[index])/periods
return df_ATR
def getBeta(df, stock_name, benchmark_name):
# Compute returns of stock
rs = roc(df[stock_name], periods=1)/100
rb = roc(df[benchmark_name], periods=1)/100
# Beta = Covariance(rs, rb)/Variance(rb)
# where rs is the return on the stock and rb is the return on a benchmark index.
return rs.cov(rb)/rb.var()
def percent_KD(df, periods=14):
"""
%K = (Current Close - Lowest Low)/(Highest High - Lowest Low) * 100
%D = 3-day SMA of %K
Lowest Low = lowest low for the look-back period
Highest High = highest high for the look-back period
%K is multiplied by 100 to move the decimal point two places
"""
current = None
if "ADJ CLOSE" in df.columns:
current = df['ADJ CLOSE']
else:
current = df['CLOSE'] # Current Close
low = df['LOW'] # Lowest Low
high = df['HIGH'] # Highest High
finish = len(df)-periods + 1
K_ = pd.DataFrame(index=df.index, columns=['%K'])
K_[0:] = np.float('nan')
for index in range(0, finish):
Highest_High = np.max(high[index: index+periods])
Lowest_Low = np.min(low[index: index+periods])
position = index + periods - 1
current_close = current.iloc[position]
K_.iloc[position] = (current_close - Lowest_Low) /(Highest_High - Lowest_Low) *100
D_ = sma(K_, periods=3)
D_.rename(columns={'%K':'%D'},inplace=True)
resultDf = K_.join(D_)
return resultDf
def OBV(df_volume, df_close):
# create empty Data Frame
df_OBV = pd.DataFrame(index = df_volume.index, columns = df_volume.columns)
# first OBV
df_OBV.iloc[0] = df_volume.iloc[0]
# Price change, df_price_change = df_close - df_close.shift(1)
df_price_change = df_close.diff(1)
for symbol in df_volume.columns:
for index in range(1, len(df_volume)):
#If the closing price is above the prior close price then:
#Current OBV = Previous OBV + Current Volume
#If the closing price is below the prior close price then:
#Current OBV = Previous OBV - Current Volume
#If the closing prices equals the prior close price then:
#Current OBV = Previous OBV (no change)
change = df_price_change.iloc[index][symbol]
current_volume = df_volume.iloc[index][symbol]
if change > 0:
current_volume = current_volume
elif change < 0:
current_volume = -1 * current_volume
else:
current_volume = 0
df_OBV.iloc[index][symbol] = df_OBV.iloc[index -1][symbol] + current_volume
return df_OBV
def change_volume(df_volume):
return df_volume.diff(1)
def get_change_price(df_open, df_close):
current_open = df_open
current_close = df_close
previous_open = df_open.shift(1)
previous_close = df_close.shift(1)
method1 = current_close - current_open
method2 = current_open - previous_open
method3 = current_open - previous_close
method4 = current_close - previous_open
method5 = current_close - previous_close
df_result = pd.concat([method1, method2, method3, method4, method5]
, axis=1
, join='inner')
df_result.columns = ['<CLOSE_OPEN>'
, '<OPEN_POPEN>'
, '<OPEN_PCLOSE>'
, '<CLOSE_POPEN>'
, '<CLOSE_PCLOSE>']
return df_result
def gain(df):
gain = df.diff(axis='columns')
return gain.sum() / df['<BUY>'].sum() * 100
import matplotlib.pyplot as plt
def compare_stock(df, x_stock, y_stock):
df_daily_returns = daily_returns(df)
df_daily_returns.plot(kind='scatter', x = x_stock, y = y_stock)
X = df_daily_returns[x_stock]
Y = df_daily_returns[y_stock]
beta, alpha = np.polyfit(X, Y, 1)
print('Beta is {}, Alpha is {}'.format(beta, alpha))
print('fx = {}x + {}'.format(beta, alpha))
#fx = beta*X + alpha
#plt.plot(X, fx, 'r-')
#plt.show()
def isUpTrend(df_price , symbol, periods =14):
finish = len(df_price) - periods +1
resultDf = pd.DataFrame(columns=df_price.columns) # empty
for index in range(1, finish):
sliced = df_price.iloc[index-1: index+periods] # slice n periods
# compute RSI
result = rsi(pd.DataFrame(sliced, columns=df_price.columns))
result = result.shift(-periods)
result.dropna(0, inplace=True)
compare = result >50 # RSI > 50
resultDf = resultDf.append(compare)
return resultDf
def listLowVolatility(df):
allDaily = daily_returns(df)
mean = allDaily.mean()
std = allDaily.std()
compare = np.abs(allDaily - mean ) > std
assert compare.shape == allDaily.shape
countVolatility = compare.sum()
indexList= np.argsort(countVolatility) # index of min values is at first of the queue
columName = allDaily.columns.values
symbol = [columName[index] for index in indexList]
# order symbol names from low variant to high variant
return symbol
def compute_gain(df, signal):
sum_buy = 0
sum_sell = 0
temp_pred = signal[0]
if temp_pred == 1:
sum_buy += df.values[0][0]
df_len = len(df)
for index in range(1, df_len):
pred = signal[index]
close = df.values[index][0]
if temp_pred == pred:
if index == df_len - 1 and pred == 1:
sum_sell += close
continue
temp_pred = pred
if pred == 0:
sum_sell += close
continue
# if pred == 1
sum_buy += close
# Gain(%) = 100 x Sum(Sell(i) - Buy(i))/Sum(Buy(i))
gain = 100 * (sum_sell - sum_buy)/sum_buy
return gain