forked from AmrutaIdagunji/Time-Series-Analysis
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Time Series Analysis - Views.py
261 lines (150 loc) · 5.47 KB
/
Time Series Analysis - Views.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
#!/usr/bin/env python
# coding: utf-8
# In[1]:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
get_ipython().run_line_magic('matplotlib', 'inline')
from matplotlib.pylab import rcParams
from datetime import datetime
rcParams['figure.figsize'] = 10, 6
# In[2]:
dataset = pd.read_csv('Views.csv', encoding='utf-16', sep = '\t')
dataset = dataset.iloc[:, 2:4]
dataset['Date'] = pd.to_datetime(dataset['Date'], infer_datetime_format = True)
indexedDataset = dataset.set_index(['Date'])
# In[3]:
#dataset['Date']
indexedDataset.tail()
# In[4]:
plt.xlabel('Date')
plt.ylabel('Total Views')
plt.plot(indexedDataset)
# In[16]:
# Find the rolling statistics
rollingMean = indexedDataset.rolling(window = 12).mean()
rollingSTD = indexedDataset.rolling(window = 12).std()
print(rollingMean, rollingSTD)
# In[17]:
# Plot Rolling Stats
orig = plt.plot(indexedDataset, color = 'blue', label='Original')
mean = plt.plot(rollingMean, color = 'green', label='Rolling Mean')
std = plt.plot(rollingSTD, color='red', label='Rolling STD')
plt.title('Rolling Stats')
plt.show(block=False)
# In[18]:
# Perform Dickey-Fuller Test
from statsmodels.tsa.stattools import adfuller
print('Results of Dickey-Fuller Test:')
dftest = adfuller(indexedDataset['# Analytics Viewed'], autolag = 'AIC')
dfoutput = pd.Series(dftest[0:4], index = ['Test Statistic', 'p-value', '#Lags Used', 'Number of Observations Used'])
for key,value in dftest[4].items():
dfoutput['Critical Value (%s)'%key] = value
print(dfoutput)
# In[19]:
# Estimate the trend
indexedDataset_logScale = np.log10(indexedDataset)
plt.plot(indexedDataset_logScale)
# In[20]:
movingAverage = indexedDataset_logScale.rolling(window = 12).mean()
movingSTD = indexedDataset_logScale.rolling(window = 12).std()
plt.plot(indexedDataset_logScale)
plt.plot(movingAverage, color='red')
plt.plot(movingSTD, color = 'black')
# In[21]:
def test_stationary(timeseries):
movingAverage = timeseries.rolling(window = 12).mean()
movingSTD = timeseries.rolling(window = 12).std()
#plot rolling stats:
orig = plt.plot(timeseries, color = 'blue', label='Original')
mean = plt.plot(movingAverage, color = 'green', label='Rolling Mean')
std = plt.plot(movingSTD, color='red', label='Rolling STD')
plt.title('Rolling Stats')
plt.show(block=False)
#Perform Dickey-Fuller test:
print('Results of Dickey-Fuller Test:')
dftest = adfuller(timeseries['# Analytics Viewed'], autolag = 'AIC')
dfoutput = pd.Series(dftest[0:4], index = ['Test Statistic', 'p-value', '#Lags Used', 'Number of Observations Used'])
for key,value in dftest[4].items():
dfoutput['Critical Value (%s)'%key] = value
print(dfoutput)
# In[22]:
datasetLDShifting = indexedDataset_logScale - indexedDataset_logScale.shift()
plt.plot(datasetLDShifting)
# In[23]:
datasetLDShifting.dropna(inplace = True)
test_stationary(datasetLDShifting)
# In[24]:
from statsmodels.tsa.seasonal import seasonal_decompose
decomposition = seasonal_decompose(indexedDataset_logScale)
trend = decomposition.trend
seasonal = decomposition.seasonal
residual = decomposition.resid
plt.subplot(411)
plt.plot(indexedDataset_logScale, label = 'Original')
plt.legend(loc = 'best')
plt.subplot(412)
plt.plot(trend, label='Trend')
plt.legend(loc='best')
plt.subplot(413)
plt.plot(seasonal, label='Seasonality')
plt.legend(loc='best')
plt.subplot(414)
plt.plot(residual, label='Residuals')
plt.legend(loc='best')
plt.tight_layout()
decomposedLogData = residual
decomposedLogData.dropna(inplace=True)
test_stationary(decomposedLogData)
# In[39]:
from statsmodels.tsa.arima_model import ARIMA
#AR Model
model = ARIMA(indexedDataset_logScale, order=(2,1,2))
results_AR = model.fit(disp = -1)
plt.plot(datasetLDShifting)
plt.plot(results_AR.fittedvalues, color='red')
plt.title('RSS: %.4f' % sum((results_AR.fittedvalues-datasetLDShifting['# Analytics Viewed'])**2))
print('Plotting AR Model')
# In[40]:
#MA Model
model = ARIMA(indexedDataset_logScale, order=(2,1,2))
results_MA = model.fit(disp = -1)
plt.plot(datasetLDShifting)
plt.plot(results_MA.fittedvalues, color='red')
plt.title('RSS: %.4f' % sum((results_MA.fittedvalues-datasetLDShifting['# Analytics Viewed'])**2))
print('Plotting MA Model')
# In[41]:
model = ARIMA(indexedDataset_logScale, order=(2,1,2))
results_ARIMA = model.fit(disp = -1)
plt.plot(datasetLDShifting)
plt.plot(results_AR.fittedvalues, color='red')
plt.title('RSS: %.4f' % sum((results_ARIMA.fittedvalues-datasetLDShifting['# Analytics Viewed'])**2))
print('Plotting ARIMA Model')
# In[42]:
predictions_ARIMA_diff = pd.Series(results_ARIMA.fittedvalues, copy=True)
print(predictions_ARIMA_diff.head())
# In[43]:
#Convert to cummulative sum
preductions_ARIMA_diff_cumsum = predictions_ARIMA_diff.cumsum()
print(preductions_ARIMA_diff_cumsum.head())
# In[44]:
predictions_ARIMA_log = pd.Series(indexedDataset_logScale['# Analytics Viewed'].ix[0], index=indexedDataset_logScale.index)
predictions_ARIMA_log = predictions_ARIMA_log.add(preductions_ARIMA_diff_cumsum, fill_value=0)
predictions_ARIMA_log.head()
# In[45]:
predictions_ARIMA= np.exp(predictions_ARIMA_log)
plt.plot(indexedDataset)
plt.plot(predictions_ARIMA)
# In[46]:
#indexedDataset_logScale.head() - 113rows
results_ARIMA.plot_predict(1, 125)
x=results_ARIMA.forecast(steps=12)
# In[47]:
results_ARIMA.forecast(steps = 120)
# In[48]:
results = results_ARIMA.forecast(steps = 12)
res=[]
for r in results:
res.append(10**r)
# In[49]:
print(res[0])