-
Notifications
You must be signed in to change notification settings - Fork 14
/
Finance.py
238 lines (223 loc) · 9.53 KB
/
Finance.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
import statistics
import numpy as np
from sklearn import linear_model
import multiprocessing
class FinanceCalculator:
def __init__(self, seriesSoFar=None, rsiPeriod=14, n_jobs=None):
self.upward = []
self.averageUpward = []
self.downward = []
self.averageDownward = []
self.rsiPeriod = rsiPeriod
if seriesSoFar is not None:
for i in range(1, len(seriesSoFar) + 1):
self.RSI(seriesSoFar[0:i])
self.lastEMA = {}
self.prevDifferences = {}
self.prevSignal = None
self.highs = []
self.lows = []
self.pKs = {}
self.pdms = []
self.ndms = []
self.trs = []
self.closes = []
self.pdisMinusNdis = []
self.obvs = {}
self.adjCloses = []
if n_jobs is None:
if multiprocessing.cpu_count() - 2 > 0:
self.jobs = multiprocessing.cpu_count() - 2
else:
self.jobs = 1
else:
self.jobs = n_jobs
def reset(self):
self.__init__()
def smaPDiff(self, series, period):
if len(series) >= period:
sma = sum(series[len(series) - period: len(series)]) / period
close = series[-1]
return ((sma - close) / close) * 100
else:
return None
def pDiffBetweenSMAs(self, series, periods):
SMAs = []
for period in periods:
if len(series) >= period:
SMAs.append(sum(series[len(series) - period: len(series)]) / period)
else:
SMAs.append(None)
pDiffSMAs = []
for i in range(0, len(SMAs)):
for j in range(i + 1, len(SMAs)):
if SMAs[i] is not None and SMAs[j] is not None and SMAs[j] != 0:
pDiffSMAs.append((SMAs[i] - SMAs[j]) / SMAs[j])
else:
pDiffSMAs.append(None)
return pDiffSMAs
# The equation for RSI can vary, but I am using the one from this video https://www.youtube.com/watch?v=WZbOeFsSirM
def RSI(self, series):
period = self.rsiPeriod # These first two variables just help readability of the code
lenSeries = len(series)
if lenSeries >= 2:
if series[lenSeries - 1] > series[lenSeries - 2]:
self.upward.append(series[lenSeries - 1] - series[lenSeries - 2])
self.downward.append(0)
elif series[lenSeries - 1] < series[lenSeries - 2]:
self.downward.append(series[lenSeries - 2] - series[lenSeries - 1])
self.upward.append(0)
else:
self.upward.append(0)
self.downward.append(0)
if lenSeries > period:
if len(self.averageUpward) == 0:
self.averageUpward.append(sum(self.upward[len(self.upward) - period:len(self.upward)]) / period)
self.averageDownward.append(
sum(self.downward[len(self.downward) - period:len(self.downward)]) / period)
else:
self.averageUpward.append((self.averageUpward[-1] * (period - 1) + self.upward[-1]) / period)
self.averageDownward.append((self.averageDownward[-1] * (period - 1) + self.downward[-1]) / period)
if self.averageDownward[-1] == 0:
return 100
else:
relativeStrength = self.averageUpward[-1] / self.averageDownward[-1]
return 100 - (100 / (relativeStrength + 1))
else:
return None
def bollingerBandsPDiff(self, series, n=20, k=2):
if len(series) >= n:
close = series[-1]
period = series[len(series) - n: len(series)]
sman = sum(period) / n
periodStd = statistics.pstdev(period)
upperBand = sman + periodStd * k
lowerBand = sman - periodStd * k
pDiffCloseUpperBand = ((upperBand - close) / close) * 100
pDiffCloseLowerBand = ((lowerBand - close) / close) * 100
pDiffSmaAbsBand = ((upperBand - sman) / sman) * 100
return pDiffCloseUpperBand, pDiffCloseLowerBand, pDiffSmaAbsBand
else:
return None, None, None
def EMA(self, series, period, label=""):
if len(series) >= period:
prevEMA = self.lastEMA.get(str(period) + label)
if prevEMA is None:
ema = sum(series[len(series) - period: len(series)]) / period
else:
factor = 2 / (period + 1)
ema = ((series[-1] - prevEMA) * factor) + prevEMA
self.lastEMA[str(period) + label] = ema
else:
ema = None
return ema
def MACD(self, series, slow, fast):
name = str(slow) + "," + str(fast)
if self.prevDifferences.get(name) is None:
self.prevDifferences[name] = []
if len(series) >= fast:
fastEMA = self.EMA(series, slow, "MACDfastEMA" + name)
slowEMA = self.EMA(series, fast, "MACDslowEMA" + name)
difference = fastEMA - slowEMA
self.prevDifferences[name].append(difference)
if len(self.prevDifferences[name]) >= 9:
signal = self.EMA(self.prevDifferences[name], 9, "MACDsignal" + name)
histogram = difference - signal
else:
signal = None
histogram = None
else:
difference = None
signal = None
histogram = None
return difference, signal, histogram
def updateHighLowClose(self, high, low, close, adjClose):
self.highs.append(high)
self.lows.append(low)
self.closes.append(close)
self.adjCloses.append(adjClose)
def stochasticOscilator(self, fastKPeriod, slowKPeriod):
if len(self.lows) >= fastKPeriod:
close = self.closes[-1]
l5 = min(self.lows[len(self.lows) - fastKPeriod: len(self.lows)])
h5 = max(self.highs[len(self.highs) - fastKPeriod: len(self.highs)])
if h5 - l5 != 0:
if self.pKs.get(str(fastKPeriod) + "," + str(slowKPeriod)) is None:
self.pKs[str(fastKPeriod) + "," + str(slowKPeriod)] = []
pK = ((close - l5) / (h5 - l5)) * 100
self.pKs[str(fastKPeriod) + "," + str(slowKPeriod)].append(pK)
if len(self.pKs[str(fastKPeriod) + "," + str(slowKPeriod)]) >= slowKPeriod:
pD = sum(self.pKs[str(fastKPeriod) + "," + str(slowKPeriod)][
len(self.pKs[str(fastKPeriod) + "," + str(slowKPeriod)]) - slowKPeriod: len(
self.pKs[str(fastKPeriod) + "," + str(slowKPeriod)])]) / slowKPeriod
else:
pD = None
else:
self.pKs[str(fastKPeriod) + "," + str(slowKPeriod)] = []
pK = None
pD = None
else:
pK = None
pD = None
return pK, pD
def ADX(self):
pdi = None
ndi = None
adx = None
if len(self.closes) >= 2:
self.trs.append(max(self.highs[-1] - self.lows[-1], abs(self.highs[-1] - self.closes[-2]),
abs(self.lows[-1] - self.closes[-2])))
moveUp = self.highs[-1] - self.highs[-2]
moveDown = self.lows[-2] - self.lows[-1]
if moveUp > moveDown and moveUp > 0:
pdm = moveUp
else:
pdm = 0
self.pdms.append(pdm)
if moveDown > moveUp and moveDown > 0:
ndm = moveDown
else:
ndm = 0
self.ndms.append(ndm)
if len(self.pdms) >= 14:
pdmEMA = self.EMA(self.pdms, 14, "ADXpdm")
ndmEMA = self.EMA(self.ndms, 14, "ADXndm")
atr = self.EMA(self.trs, 14, "ADXatr")
pdi = 100 * (pdmEMA / atr)
ndi = 100 * (ndmEMA / atr)
self.pdisMinusNdis.append(abs(pdi - ndi))
if len(self.pdisMinusNdis) >= 14:
adx = 100 * self.EMA(self.pdisMinusNdis, 14) / (pdi + ndi)
return pdi, ndi, adx
def OBVGrad(self, volume, period):
if self.obvs.get(period) is None:
self.obvs[period] = [0]
if len(self.adjCloses) >= 2:
adjClosePChange = ((self.adjCloses[-1] - self.adjCloses[-2]) / self.adjCloses[-2]) * 100
else:
return None
if adjClosePChange > 0:
obv = self.obvs[period][-1] + volume
elif adjClosePChange < 0:
obv = self.obvs[period][-1] - volume
elif adjClosePChange == 0:
obv = self.obvs[period][-1]
self.obvs[period].append(obv)
OBVGrad = None
if len(self.obvs[period]) > period:
OBVSample = self.obvs[period][len(self.obvs[period]) - period: len(self.obvs[period])]
OBVSample = np.array(OBVSample)
x = np.arange(1, period + 1).reshape(period, 1)
regr = linear_model.LinearRegression(n_jobs=self.jobs)
regr.fit(x, OBVSample)
OBVGrad = regr.coef_[0]
return OBVGrad
def adjCloseGrad(self, period):
adjCloseGrad = None
if len(self.adjCloses) > period:
adjCloseSample = np.array(self.adjCloses[len(self.adjCloses) - period: len(self.adjCloses)])
x = np.arange(1, period + 1).reshape(period, 1)
regr = linear_model.LinearRegression(n_jobs=self.jobs)
regr.fit(x, adjCloseSample)
adjCloseGrad = regr.coef_[0]
return adjCloseGrad