-
Notifications
You must be signed in to change notification settings - Fork 1
/
daily_quotes_analysis.py
388 lines (302 loc) · 26.7 KB
/
daily_quotes_analysis.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
### INITIALIZING THE REQUIRED LIBRARIES
import yfinance as yf
import pandas as pd
from datetime import datetime, timedelta, date
from dateutil.relativedelta import relativedelta
import numpy as np
import os
import pandas_gbq
import pyarrow
# Reference: https://www.datacamp.com/community/tutorials/moving-averages-in-pandas
### CAPTURE CURRENT HOLDINGS IN A LIST ###
current_local = os.getcwd()
if current_local[0:2] == 'C:':
daily_data_holdings = "VOO VTI XITK IHI ARKG"
else:
daily_data_holdings = "VTI FZROX FSKAX VOO IVV FXAIX FNILX VGT FTEC QQQ XITK VHT FHLC IHI XHE VYM SCHD FPBFX FIVFX SMH XSD ARKK ARKW ARKF ARKQ ARKG WCLD SKYY SLV GLDM IAU BND AGG FNBGX WFC TSLA FSCSX FSELX FSPHX FBIOX FFNOX AAPL"
### CATEGORIZING STOCKS
investment_account = ['VTI', 'FZROX', 'FSKAX', 'VOO', 'IVV', 'FXAIX', 'FNILX', 'VGT', 'FTEC', 'QQQ', 'XITK', 'VHT', 'FHLC', 'IHI', 'XHE', 'VYM', 'SCHD', 'FPBFX', 'FIVFX', 'SMH', 'XSD', 'ARKK', 'ARKW', 'ARKF', 'ARKQ', 'ARKG', 'WCLD', 'SKYY', 'SLV', 'GLDM', 'IAU', 'BND', 'AGG', 'FNBGX', 'WFC', 'TSLA', 'AAPL']
retirement_account = ['FZROX', 'FSKAX', 'FXAIX', 'FNILX', 'FSCSX', 'FSELX', 'FSPHX', 'FBIOX', 'FFNOX']
s_and_p_500 = ['FXAIX', 'VOO', 'IVV', 'FNILX']
total_market = ['VTI', 'FZROX', 'FSKAX', 'VOO']
technology = ['VGT', 'FTEC', 'XITK', 'FSCSX', 'QQQ']
semiconductors = ['SMH', 'XSD', 'FSELX']
dividends = ['VYM', 'SCHD', 'WFC']
health = ['VHT', 'FHLC', 'IHI', 'XHE', 'FSPHX', 'FBIOX']
high_growth = ['ARKK', 'ARKW', 'ARKF', 'ARKQ', 'ARKG', 'WCLD', 'SKYY', 'TSLA', 'AAPL']
international = ['FPBFX', 'FIVFX', 'FFNOX']
metals = ['SLV', 'GLDM', 'IAU']
bond = ['BND', 'AGG', 'FNBGX']
### PREPARING DATA FOR DAILY AVERAGE ###
start_date = str(int(datetime.today().strftime('%Y')) - 12) + '-' + datetime.today().strftime('%m') + '-' + datetime.today().strftime('%d') # Pulling 10 Years of data
daily_df = yf.download(daily_data_holdings, start=start_date, interval="1d")
daily_df = daily_df.unstack().reset_index()
daily_df.columns = ['col_name', 'Ticker', 'Date', 'Value']
daily_df = daily_df.pivot_table(values='Value', index=['Ticker', 'Date'],columns='col_name').reset_index()
daily_df = daily_df.sort_values(['Ticker', 'Date'], ascending=False)
daily_df.columns.name = None
col_names = ['ticker', 'date', 'adj_close', 'close', 'high', 'low', 'open', 'volume']
daily_df.columns = col_names
daily_df = daily_df.sort_values(by=['ticker', 'date'], axis=0, ascending=True, kind='mergesort').reset_index(drop=True) #mergesort works better on pre-sorted items. For most other cases, quicksort works good
daily_df['account'] = np.where((daily_df.ticker.isin(investment_account)) & (daily_df.ticker.isin(retirement_account)), "Both Accounts", np.where(daily_df.ticker.isin(investment_account), "Investment Account", "Retirement Account"))
daily_df['sector'] = np.where(daily_df.ticker.isin(s_and_p_500),'S&P 500', np.where(daily_df.ticker.isin(total_market),'Total Market', np.where(daily_df.ticker.isin(technology),'Technology', np.where(daily_df.ticker.isin(semiconductors),'Semiconductors', np.where(daily_df.ticker.isin(health),'Health Service & Devices', np.where(daily_df.ticker.isin(dividends),'Dividends', np.where(daily_df.ticker.isin(high_growth),'High Growth', np.where(daily_df.ticker.isin(international),'International', np.where(daily_df.ticker.isin(bond),'Bonds', np.where(daily_df.ticker.isin(metals),'Precious Metals','Not Applicable'))))))))))
stock_data_analysis = daily_df.copy()
### APPLYING FILTERS TO THE DATASET
summary_date_start_filter = int(datetime.today().strftime('%Y')) - 10 # Filter 10 Years from CY-1
summary_date_end_filter = int(datetime.today().strftime('%Y'))-1 # CY-1
stock_data_analysis = stock_data_analysis[stock_data_analysis['date'].dt.year.isin(list(range(summary_date_start_filter, summary_date_end_filter+1)))]
stock_date_filter = str(int(datetime.today().strftime('%Y')) - 10) + '-' + datetime.today().strftime('%m') + '-' + datetime.today().strftime('%d')
daily_df = daily_df[daily_df['date'] >= stock_date_filter]
price_prediction_data = daily_df[['ticker', 'date', 'adj_close', 'high', 'low']].copy().reset_index(drop=True)
# CALCULATING MOVING AVERAGES
# daily_df['sma_50'] = daily_df.groupby(['ticker'])['adj_close'].rolling(window=50).mean().reset_index(drop=True)
# daily_df['sma_100'] = daily_df.groupby(['ticker'])['adj_close'].rolling(window=100).mean().reset_index(drop=True)
# daily_df['sma_200'] = daily_df.groupby(['ticker'])['adj_close'].rolling(window=200).mean().reset_index(drop=True)
# CALCULATING OTHER CALCULATED COLUMNS
daily_df['daily_change_pct'] = (daily_df.adj_close - daily_df.groupby(['ticker']).adj_close.shift(1)) / daily_df.adj_close
daily_df = daily_df.sort_values(by=['ticker', 'date'], axis=0, ascending=True, kind='mergesort').reset_index(drop=True)
# CALCULATING CHANGE FROM 52 WEEKS
daily_df['52_wk_comparison_price'] = np.where(daily_df.date.dt.year % 4 == 0, daily_df.groupby(['ticker']).adj_close.shift(366), daily_df.groupby(['ticker']).adj_close.shift(365))
daily_df['first_value'] = daily_df.groupby(['ticker'])['adj_close'].transform('first').reset_index(drop=True)
daily_df['52_wk_comparison_price'] = np.where((daily_df['52_wk_comparison_price'].isnull()) & (daily_df.date == daily_df.date.max()), daily_df['first_value'], daily_df['52_wk_comparison_price'])
daily_df['change_since_52_weeks'] = (daily_df['adj_close'] - daily_df['52_wk_comparison_price']) / daily_df.adj_close
# CALCULATING RSI
# LET US CALCULATE THE RSI USING THE PERIOD = 28
# RSI CALCULATION REFERENCE - https://www.macroption.com/rsi-calculation/
period = 28 # Initializing Period to calculate RSI
daily_df['ups'] = np.where((daily_df.adj_close.isnull()) | (daily_df.groupby(['ticker']).adj_close.shift(1).isnull()) , np.nan, np.where(daily_df.adj_close - daily_df.groupby(['ticker']).adj_close.shift(1) > 0, daily_df.adj_close - daily_df.groupby(['ticker']).adj_close.shift(1), 0))
daily_df['downs'] = np.where((daily_df.adj_close.isnull()) | (daily_df.groupby(['ticker']).adj_close.shift(1).isnull()) , np.nan, np.where(daily_df.adj_close - daily_df.groupby(['ticker']).adj_close.shift(1) < 0, abs(daily_df.adj_close - daily_df.groupby(['ticker']).adj_close.shift(1)), 0))
daily_df['ups_avg'] = np.where((daily_df.ups.isnull()) | (daily_df.downs.isnull()), np.nan, daily_df.groupby(['ticker'])['ups'].rolling(window=period).mean().reset_index(drop=True).shift(1))
daily_df['downs_avg'] = np.where((daily_df.ups.isnull()) | (daily_df.downs.isnull()), np.nan, daily_df.groupby(['ticker'])['downs'].rolling(window=period).mean().reset_index(drop=True).shift(1))
daily_df['relative_strength_index'] = 100 - (100 / (1 + (daily_df.ups_avg / daily_df.downs_avg)))
daily_df = daily_df.sort_values(by=['ticker', 'date'], axis=0, ascending=True, kind='mergesort').reset_index(drop=True)
# CHECK THE FIRST DATA POINT FOR A TICKER - daily_df[1480:1521]
# CALCULATE MACD ON TOP OF ADJUSTED CLOSING PRICE
# CALCULATION REFERENCE - https://www.youtube.com/watch?v=9wqvjl_smv4&t=14s&ab_channel=Troy%26Vaishali
# CALCULATION REFERENCE - https://www.investopedia.com/ask/answers/122414/what-moving-average-convergence-divergence-macd-formula-and-how-it-calculated.asp
# PYTHON REFERENCE - https://towardsdatascience.com/trading-toolbox-02-wma-ema-62c22205e2a9
faster_period = 12
slower_period = 26
signal_period = 9
daily_df['ema_12'] = daily_df.groupby(['ticker'])['close'].transform(lambda x: x.ewm(span=faster_period, adjust=False).mean().reset_index(drop=True))
daily_df['ema_26'] = daily_df.groupby(['ticker'])['close'].transform(lambda x: x.ewm(span=slower_period, adjust=False).mean().reset_index(drop=True))
daily_df['macd_line'] = daily_df['ema_12'] - daily_df['ema_26']
daily_df['signal_line'] = daily_df.groupby(['ticker'])['macd_line'].transform(lambda x: x.ewm(span=signal_period, adjust=False).mean().reset_index(drop=True))
daily_df['macd_histogram'] = daily_df['macd_line'] - daily_df['signal_line']
# SUBSETTING 10 YEARS DATA AND CREATING MAX AND MIN DATE FILTER
# daily_df = daily_df[daily_df.date.dt.year >= daily_df.date.dt.year.max()-10].reset_index(drop=True)
daily_df['date_filter'] = np.where(daily_df.date == daily_df.groupby('ticker').date.transform('max').reset_index(drop=True),"MAX", "NO")
# SUBSETTING THE COLUMNS TO KEEP
cols_to_keep = ['ticker', 'date', 'adj_close', 'close', 'high', 'low', 'open', 'volume', 'account', 'sector', 'daily_change_pct', 'date_filter', 'relative_strength_index', 'change_since_52_weeks', 'ema_12', 'ema_26', 'macd_line', 'signal_line', 'macd_histogram']
daily_df = daily_df[cols_to_keep]
# PREPARING THE STOCK ANALYSIS DATA
# Date Open High Low Close Volume Dividends Stock Splits Stock Adj Close Year Month
# ticker open high low close volume ticker adj_close year month
stock_data_analysis['year'] = stock_data_analysis.date.dt.year
stock_data_analysis['month'] = stock_data_analysis['date'].dt.month_name().str.slice(stop=3)
# ## TOP MONTHS TO SELL THE STOCK
temp = stock_data_analysis[['high', 'ticker', 'year', 'month']].copy().reset_index(drop=True)
temp['mean_highs'] = temp.groupby(['ticker', 'year', 'month'])['high'].transform('mean').reset_index(drop=True)
temp['rank'] = temp.groupby(['ticker', 'year'])['mean_highs'].rank(method='dense', ascending=False)
temp = temp[temp['rank'] <= 3].sort_values(by=['ticker', 'year', 'rank'], ascending=False)
temp = temp[['ticker', 'year', 'month', 'rank']].drop_duplicates().reset_index(drop=True)
temp['month_count'] = temp.groupby(['ticker', 'month'])['year'].transform('count')
temp = temp.pivot_table(index=['ticker', 'month', 'rank', 'month_count'], aggfunc={'year': 'count'}).sort_values(by=['ticker', 'rank', 'year'], ascending=[True, True, False]).reset_index()
temp['rank1'] = temp.groupby(['ticker', 'rank'])['year'].rank(method='dense', ascending=False)
temp = temp[temp['rank1'] == 1].copy().reset_index(drop=True)
temp['months_to_sell'] = temp['month'] + " (" + temp['month_count'].astype(int).astype(str) + ", " + temp['year'].astype(int).astype(str) + ")"
months_to_sell = temp.groupby(['ticker'])['months_to_sell'].apply(', '.join).reset_index()
### TOP MONTHS TO BUY THE STOCKS
temp = stock_data_analysis[['low', 'ticker', 'year', 'month']].copy().reset_index(drop=True)
temp['mean_lows'] = temp.groupby(['ticker', 'year', 'month'])['low'].transform('mean').reset_index(drop=True)
temp['rank'] = temp.groupby(['ticker', 'year'])['mean_lows'].rank(method='dense', ascending=True)
temp = temp[temp['rank'] <= 3].sort_values(by=['ticker', 'year', 'rank'], ascending=True)
temp = temp[['ticker', 'year', 'month', 'rank']].drop_duplicates()
temp['month_count'] = temp.groupby(['ticker', 'month'])['year'].transform('count')
temp = temp.pivot_table(index=['ticker', 'month', 'rank', 'month_count'], aggfunc={'year': 'count'}).sort_values(by=['ticker', 'rank', 'year'], ascending=[True, True, False]).reset_index()
temp['rank1'] = temp.groupby(['ticker', 'rank'])['year'].rank(method='dense', ascending=False)
temp = temp[temp['rank1'] == 1].copy().reset_index(drop=True)
temp['months_to_buy'] = temp['month'] + " (" + temp['month_count'].astype(int).astype(str) + ", " + temp['year'].astype(int).astype(str) + ")"
months_to_buy = temp.groupby(['ticker'])['months_to_buy'].apply(', '.join).reset_index()
### GENERATE GROWTH METRICS FOR THE STOCKS
temp = stock_data_analysis[['adj_close', 'ticker', 'year', 'month', 'date']].loc[stock_data_analysis['month']=="Dec"].sort_values(by=['ticker', 'year']).reset_index().copy()
temp['max_dec'] = temp.groupby(['ticker', 'year']).date.transform('max').reset_index(drop=True)
temp = temp[temp['date']==temp['max_dec']][['ticker', 'year', 'month', 'adj_close']].copy()
temp['lagged_close'] = temp.groupby(by=['ticker'])['adj_close'].shift(1)
temp['yearly_growth'] = (temp['adj_close'] - temp['lagged_close']) / temp['lagged_close']
avg_yearly_growth = temp.groupby('ticker')['yearly_growth'].mean().reset_index()
temp['growth_probability'] = np.where(np.isnan(temp['yearly_growth']), np.NaN, np.where(temp['yearly_growth'] > 0,1,-1))
growth_tendency = (temp.groupby('ticker')['growth_probability'].sum() / (temp.groupby('ticker')['growth_probability'].count())).reset_index()
growth_metrics = pd.merge(avg_yearly_growth, growth_tendency, how="inner")
### CALCUALTE THE TOTAL HISTORICAL DATA POINTS
temp = stock_data_analysis[['ticker', 'date']].copy().reset_index(drop=True)
temp['max_date'] = temp.groupby('ticker')['date'].transform('max').reset_index(drop=True)
temp['min_date'] = temp.groupby('ticker')['date'].transform('min').reset_index(drop=True)
temp = temp[['ticker', 'max_date', 'min_date']].copy().drop_duplicates()
temp['no_of_months'] = ((temp.max_date - temp.min_date) / np.timedelta64(1, 'M')).round(decimals=0).astype('int')
total_data_points = temp[['ticker', 'no_of_months']].copy()
### SUBSETTING 1 YR DATA TO GENERATE 52 WEEK METRICS
temp = price_prediction_data.copy().reset_index(drop=True)
max_period = temp.date.max()
filter_period = np.where(max_period.year % 4 == 0, max_period - timedelta(days=366), max_period - timedelta(days=365))
temp = temp[temp.date >= filter_period]
temp['52_week_high'] = temp.groupby('ticker')['high'].transform('max')
temp['52_week_low'] = temp.groupby('ticker')['low'].transform('min')
temp['52_week_mean'] = temp.groupby('ticker')['adj_close'].transform('mean')
temp['52_week_median'] = temp.groupby('ticker')['adj_close'].transform('median')
fifty_two_week_metric = temp[['ticker', '52_week_high', '52_week_low', '52_week_mean', '52_week_median']].drop_duplicates().reset_index(drop=True)
### CODE FOR STOCK PRICE PREDICTION
price_prediction_data['year'] = price_prediction_data.date.dt.year
price_prediction_data['month'] = price_prediction_data['date'].dt.month
price_prediction_data['mon'] = price_prediction_data['date'].dt.month_name().str.slice(stop=3)
temp = price_prediction_data.copy().reset_index(drop=True)
temp['median_price'] = temp.groupby(['ticker', 'year', 'month'])['adj_close'].transform('median').reset_index(drop=True)
temp = temp[['ticker', 'year', 'month', 'mon', 'median_price']].drop_duplicates().sort_values(by=(['ticker', 'year', 'month'])).reset_index(drop=True)
temp2 = temp.copy() # To be used for calculating the actual price prediction
temp['cm_12'] = temp.groupby('ticker')['median_price'].shift(12)
temp['cm_9'] = temp.groupby('ticker')['median_price'].shift(9)
temp['cm_6'] = temp.groupby('ticker')['median_price'].shift(6)
temp['cm_3'] = temp.groupby('ticker')['median_price'].shift(3)
temp['cm_12_pct'] = (temp['median_price'] - temp['cm_12']) / temp['cm_12']
temp['cm_9_pct'] = (temp['median_price'] - temp['cm_9']) / temp['cm_9']
temp['cm_6_pct'] = (temp['median_price'] - temp['cm_6']) / temp['cm_6']
temp['cm_3_pct'] = (temp['median_price'] - temp['cm_3']) / temp['cm_3']
temp['change_12_months'] = temp.groupby(['ticker', 'month'])['cm_12_pct'].transform('mean')
temp['change_9_months'] = temp.groupby(['ticker', 'month'])['cm_9_pct'].transform('mean')
temp['change_6_months'] = temp.groupby(['ticker', 'month'])['cm_6_pct'].transform('mean')
temp['change_3_months'] = temp.groupby(['ticker', 'month'])['cm_3_pct'].transform('mean')
base_pct_change_data = temp[['ticker', 'month', 'mon', 'change_12_months', 'change_9_months', 'change_6_months', 'change_3_months']].copy().sort_values(by=(['ticker', 'month'])).drop_duplicates().reset_index(drop=True)
### CALCULATING THE PREDICTED VALUES FOR ACTUAL MONTH
## CALCULATING FOR THE CURRENT MONTH
current_year_month_filter = str(int(datetime.today().strftime('%Y'))) + str(int(datetime.today().strftime('%m')))
current_dt = date(int(datetime.today().strftime('%Y')), int(datetime.today().strftime('%m')), 1)
month1_filter = np.array([int((current_dt - relativedelta(months=3)).strftime('%m')), int((current_dt - relativedelta(months=6)).strftime('%m')), int((current_dt - relativedelta(months=9)).strftime('%m')), int((current_dt - relativedelta(months=12)).strftime('%m'))])
month2_filter = np.array([int((current_dt - relativedelta(months=2)).strftime('%m')), int((current_dt - relativedelta(months=5)).strftime('%m')), int((current_dt - relativedelta(months=8)).strftime('%m')), int((current_dt - relativedelta(months=11)).strftime('%m'))])
month3_filter = np.array([int((current_dt - relativedelta(months=1)).strftime('%m')), int((current_dt - relativedelta(months=4)).strftime('%m')), int((current_dt - relativedelta(months=7)).strftime('%m')), int((current_dt - relativedelta(months=10)).strftime('%m'))])
# FIXING THE WEIGHTS FOR THE MONTHS
wt_3 = 2
wt_6 = 1
wt_9 = 1
wt_12 = 0.75
wt_list = [wt_3, wt_6, wt_9, wt_12]
temp = temp2[(temp2.year >= summary_date_end_filter) & (temp2.year.astype(str) + temp2.month.astype(str) != current_year_month_filter) & (temp2.month.isin(month1_filter))].copy()
temp['date'] = (temp['year'].astype(str) + temp['month'].astype(str).str.pad(width=2, side='left', fillchar='0')).astype(int)
temp['rank'] = temp.groupby(['ticker'])['date'].rank(method='first', ascending=False).copy()
temp = temp[temp['rank'] <= 4]
temp['month_category'] = pd.cut(temp['rank'], bins=[0,1,2,3,4], include_lowest=True, labels=['month3', 'month6', 'month9', 'month12'])
temp = temp[['ticker', 'median_price', 'month_category']]
temp = temp.pivot_table(index='ticker', columns='month_category', aggfunc={'median_price':'mean'})
temp.columns = temp.columns.droplevel(0) # This is to remove the median from the column level
temp = pd.DataFrame(temp.to_records()) # This will remove any multilevel indexes, and also convert the index to column and reset index
temp['month'] = int(datetime.today().strftime('%m'))
current_month_price = pd.merge(temp, base_pct_change_data, how='inner', on=['ticker', 'month']).reset_index(drop=True)
current_month_price['change_3_months_value'] = (current_month_price['month3']+(current_month_price['month3']*current_month_price['change_3_months']))
current_month_price['change_6_months_value'] = (current_month_price['month6']+(current_month_price['month6']*current_month_price['change_6_months']))
current_month_price['change_9_months_value'] = (current_month_price['month9']+(current_month_price['month3']*current_month_price['change_9_months']))
current_month_price['change_12_months_value'] = (current_month_price['month12']+(current_month_price['month12']*current_month_price['change_12_months']))
data = np.array(current_month_price[['change_3_months_value', 'change_6_months_value', 'change_9_months_value', 'change_12_months_value']]) # USING NUMPY AS WEIGHTED AVERAGE IS NOT AVAILABLE WITH PANDAS
masked_data = np.ma.masked_array(data, np.isnan(data)) # USING A MASKED ARRAY TO OVERCOME ISSUES WITH NA
current_month_price['cm_target_price'] = np.ma.average(masked_data, axis=1, weights=wt_list).filled(np.nan)
current_month_price = current_month_price[['ticker', 'cm_target_price']].reset_index(drop=True)
### CALCULATIONS FOR CURRENT MONTH PLUS ONE
temp = temp2[(temp2.year >= summary_date_end_filter) & (temp2.year.astype(str) + temp2.month.astype(str) != current_year_month_filter) & (temp2.month.isin(month2_filter))].copy()
temp['date'] = (temp['year'].astype(str) + temp['month'].astype(str).str.pad(width=2, side='left', fillchar='0')).astype(int)
temp['rank'] = temp.groupby(['ticker'])['date'].rank(method='first', ascending=False).copy()
temp = temp[temp['rank'] <= 4]
temp['month_category'] = pd.cut(temp['rank'], bins=[0,1,2,3,4], include_lowest=True, labels=['month3', 'month6', 'month9', 'month12'])
temp = temp[['ticker', 'median_price', 'month_category']]
temp = temp.pivot_table(index='ticker', columns='month_category', aggfunc={'median_price':'mean'})
temp.columns = temp.columns.droplevel(0) # This is to remove the median from the column level
temp = pd.DataFrame(temp.to_records()) # This will remove any multilevel indexes, and also convert the index to column and reset index
temp['month'] = int((datetime.today()+relativedelta(months=+1)).strftime('%m'))
current_month_one_price = pd.merge(temp, base_pct_change_data, how='inner', on=['ticker', 'month']).reset_index(drop=True)
current_month_one_price['change_3_months_value'] = (current_month_one_price['month3']+(current_month_one_price['month3']*current_month_one_price['change_3_months']))
current_month_one_price['change_6_months_value'] = (current_month_one_price['month6']+(current_month_one_price['month6']*current_month_one_price['change_6_months']))
current_month_one_price['change_9_months_value'] = (current_month_one_price['month9']+(current_month_one_price['month3']*current_month_one_price['change_9_months']))
current_month_one_price['change_12_months_value'] = (current_month_one_price['month12']+(current_month_one_price['month12']*current_month_one_price['change_12_months']))
data = np.array(current_month_one_price[['change_3_months_value', 'change_6_months_value', 'change_9_months_value', 'change_12_months_value']]) # USING NUMPY AS WEIGHTED AVERAGE IS NOT AVAILABLE WITH PANDAS
masked_data = np.ma.masked_array(data, np.isnan(data)) # USING A MASKED ARRAY TO OVERCOME ISSUES WITH NA
current_month_one_price['cm_plus_one_target_price'] = np.ma.average(masked_data, axis=1, weights=wt_list).filled(np.nan)
current_month_one_price = current_month_one_price[['ticker', 'cm_plus_one_target_price']].reset_index(drop=True)
### CALCULATIONS FOR CURRENT MONTH PLUS TWO
temp = temp2[(temp2.year >= summary_date_end_filter) & (temp2.year.astype(str) + temp2.month.astype(str) != current_year_month_filter) & (temp2.month.isin(month3_filter))].copy()
temp['date'] = (temp['year'].astype(str) + temp['month'].astype(str).str.pad(width=2, side='left', fillchar='0')).astype(int)
temp['rank'] = temp.groupby(['ticker'])['date'].rank(method='first', ascending=False).copy()
temp = temp[temp['rank'] <= 4]
temp['month_category'] = pd.cut(temp['rank'], bins=[0,1,2,3,4], include_lowest=True, labels=['month3', 'month6', 'month9', 'month12'])
temp = temp[['ticker', 'median_price', 'month_category']]
temp = temp.pivot_table(index='ticker', columns='month_category', aggfunc={'median_price':'mean'})
temp.columns = temp.columns.droplevel(0) # This is to remove the median from the column level
temp = pd.DataFrame(temp.to_records()) # This will remove any multilevel indexes, and also convert the index to column and reset index
temp['month'] = int((datetime.today()+relativedelta(months=+2)).strftime('%m'))
current_month_two_price = pd.merge(temp, base_pct_change_data, how='inner', on=['ticker', 'month']).reset_index(drop=True)
current_month_two_price['change_3_months_value'] = (current_month_two_price['month3']+(current_month_two_price['month3']*current_month_two_price['change_3_months']))
current_month_two_price['change_6_months_value'] = (current_month_two_price['month6']+(current_month_two_price['month6']*current_month_two_price['change_6_months']))
current_month_two_price['change_9_months_value'] = (current_month_two_price['month9']+(current_month_two_price['month3']*current_month_two_price['change_9_months']))
current_month_two_price['change_12_months_value'] = (current_month_two_price['month12']+(current_month_two_price['month12']*current_month_two_price['change_12_months']))
data = np.array(current_month_two_price[['change_3_months_value', 'change_6_months_value', 'change_9_months_value', 'change_12_months_value']]) # USING NUMPY AS WEIGHTED AVERAGE IS NOT AVAILABLE WITH PANDAS
masked_data = np.ma.masked_array(data, np.isnan(data)) # USING A MASKED ARRAY TO OVERCOME ISSUES WITH NA
current_month_two_price['cm_plus_two_target_price'] = np.ma.average(masked_data, axis=1, weights=wt_list).filled(np.nan)
current_month_two_price = current_month_two_price[['ticker', 'cm_plus_two_target_price']].reset_index(drop=True)
### MERGING ALL THE PREDICTED PRICE DATASET
price_predictions = pd.merge(current_month_price, current_month_one_price, how='left', on='ticker')
price_predictions = pd.merge(price_predictions, current_month_two_price, how='left', on='ticker')
### MERGING ALL SUMMARY DATASET
summary_data = stock_data_analysis[['ticker', 'account', 'sector']].drop_duplicates().copy()
summary_data = pd.merge(summary_data, total_data_points, on='ticker', how='inner')
summary_data = pd.merge(summary_data, growth_metrics, on='ticker', how='inner')
summary_data = pd.merge(summary_data, months_to_sell, on='ticker', how='inner')
summary_data = pd.merge(summary_data, months_to_buy, on='ticker', how='inner')
summary_data = pd.merge(summary_data, fifty_two_week_metric, on='ticker', how='inner')
summary_data = pd.merge(summary_data, price_predictions, on='ticker', how='left')
# summary_data['target_price'] = ((summary_data['52_week_high'] * 1) + (summary_data['52_week_low'] * 1) + (summary_data['52_week_mean'] * 2) + summary_data['52_week_median'] * 2.5) / 6.5
### WRITING DATA TO LOCAL DRIVE
if current_local[0:2] == 'C:':
summary_data.to_csv(current_local + "\\us-stocks-analysis\\input\\summary_data_new.csv", index=False)
else:
summary_data.to_csv("~/my-portfolio-analysis/input-files/summary_data.csv", index=False)
print("Summary data file successfully exported to local drive")
# WRITING PANDAS DATAFRAME TO BIGQUERY DATASET
pandas_gbq.to_gbq(daily_df, 'portfolio_data.daily_quotes_analysis', project_id= 'my-portfolio-analysis', if_exists='replace')
### CREATING PYTHON BIGQUERY CLIENT OBJECT
# from google.cloud import bigquery
# from google.oauth2 import service_account
# credentials = service_account.Credentials.from_service_account_file(current_local + "\\us-stocks-analysis\\input\\big-query-key\\bq-key.json")
# project_id = 'my-portfolio-analysis'
# dataset = 'portfolio_data'
# table = 'daily_quotes_test'
# client = bigquery.Client(credentials=credentials, project=project_id)
# job_config = bigquery.LoadJobConfig(
# schema = [
# bigquery.SchemaField("ticker", "STRING"),
# bigquery.SchemaField("date", "TIMESTAMP"),
# bigquery.SchemaField("adj_close", "FLOAT64"),
# bigquery.SchemaField("close", "FLOAT64"),
# bigquery.SchemaField("high", "FLOAT64"),
# bigquery.SchemaField("low", "FLOAT64"),
# bigquery.SchemaField("open", "FLOAT64"),
# bigquery.SchemaField("volume", "FLOAT64"),
# bigquery.SchemaField("account", "STRING"),
# bigquery.SchemaField("sector", "STRING"),
# bigquery.SchemaField("daily_change_pct", "FLOAT64"),
# bigquery.SchemaField("date_filter", "STRING"),
# bigquery.SchemaField("relative_strength_index", "FLOAT64"),
# bigquery.SchemaField("change_since_52_weeks", "FLOAT64"),
# bigquery.SchemaField("ema_12", "FLOAT64"),
# bigquery.SchemaField("ema_26", "FLOAT64"),
# bigquery.SchemaField("macd_line", "FLOAT64"),
# bigquery.SchemaField("signal_line", "FLOAT64"),
# bigquery.SchemaField("macd_histogram", "FLOAT64")
# ],
# time_partitioning = bigquery.TimePartitioning(
# type_=bigquery.TimePartitioningType.DAY,
# field="date"
# ),
# write_disposition = bigquery.WriteDisposition.WRITE_TRUNCATE
# )
# load_job = client.load_table_from_dataframe(
# daily_df, '.'.join([project_id, dataset, table]), job_config=job_config
# )
# result = load_job.result()
# PRINTING SUCCESSFUL CODE EXECUTION MESSAGE
print("Writing to BigQuery over daily_quotes_analysis successfully completed")