-
Notifications
You must be signed in to change notification settings - Fork 1
/
FFReg.py
541 lines (417 loc) · 23.8 KB
/
FFReg.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
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
##############################################################################
# Copyright 2020 Marco Marra <marrez89@gmail.com>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
##############################################################################
import pandas as pd
import statsmodels.api as sm
from glob import glob
import ntpath
import pandas_datareader.data as web
import os
import runcurl
from urllib.error import HTTPError
pd.options.display.max_columns = None
pd.options.display.width = None
def get_morningstar_fund_name(fund_isin):
# curl string to obtain some morningstar fund info
curlstr = "curl 'https://www.morningstar.com/api/v1/search/securities?q=" + fund_isin + "®ion=international&limit=50' -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:80.0) Gecko/20100101 Firefox/80.0' -H 'Accept: application/json, text/plain, */*' -H 'Accept-Language: en-US,en;q=0.5' --compressed -H 'Referer: https://www.morningstar.com/search?query=IE00BFY0GT14' -H 'x-api-key: Nrc2ElgzRkaTE43D5vA7mrWj2WpSBR35fvmaqfte' -H 'DNT: 1' -H 'Connection: keep-alive' -H 'Cookie: ASP.NET_SessionId=zdgbrkbblfiepa45gm4wjffg; _gcl_au=1.1.1480557771.1600715785; _uetsid=c33fb5d1faf382624c557dd389f438fb; _uetvid=11b9521be1a452a32efc4091dbb6616e; overlay_hibernation=Wed%2C%2023%20Sep%202020%2018:51:51%20GMT; intro_hibernation=Tue%2C%2022%20Sep%202020%2019:51:51%20GMT' -H 'TE: Trailers'"
req = runcurl.execute(curlstr)
# extract the morningstar fund info
if req.status_code == 200:
fund_name = req.json()['results'][0]['name']
print(fund_isin + ': ' + fund_name)
return fund_name
else:
print(fund_isin + ': cannot retrieve fund name from Morningstar!')
return ''
def get_morningstar_fund_category(fund_isin):
# curl string to obtain the morningstar fund id
curlstr = "curl 'https://www.morningstar.com/api/v1/search/securities?q=" + fund_isin + "®ion=international&limit=50' -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:80.0) Gecko/20100101 Firefox/80.0' -H 'Accept: application/json, text/plain, */*' -H 'Accept-Language: en-US,en;q=0.5' --compressed -H 'Referer: https://www.morningstar.com/search?query=IE00BFY0GT14' -H 'x-api-key: Nrc2ElgzRkaTE43D5vA7mrWj2WpSBR35fvmaqfte' -H 'DNT: 1' -H 'Connection: keep-alive' -H 'Cookie: ASP.NET_SessionId=zdgbrkbblfiepa45gm4wjffg; _gcl_au=1.1.1480557771.1600715785; _uetsid=c33fb5d1faf382624c557dd389f438fb; _uetvid=11b9521be1a452a32efc4091dbb6616e; overlay_hibernation=Wed%2C%2023%20Sep%202020%2018:51:51%20GMT; intro_hibernation=Tue%2C%2022%20Sep%202020%2019:51:51%20GMT' -H 'TE: Trailers'"
req = runcurl.execute(curlstr)
# extract the morningstar fund id
if req.status_code == 200:
if not req.json()['results']:
print(fund_isin + ': cannot retrieve fund data from Morningstar!')
return ''
else:
sec_id = req.json()['results'][0]['secId']
else:
print(fund_isin + ': cannot retrieve fund data from Morningstar!')
return ''
# curl string to obtain the morningstar fund info
curlstr = "curl 'https://api-global.morningstar.com/sal-service/v1/etf/process/asset/v2/" + sec_id + "/data?locale=en&clientId=MDC&benchmarkId=category&version=3.31.0' -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:81.0) Gecko/20100101 Firefox/81.0' -H 'Accept: application/json, text/plain, */*' -H 'Accept-Language: en-US,en;q=0.5' --compressed -H 'Referer: https://www.morningstar.com/etfs/xetr/exsa/portfolio' -H 'X-API-REALTIME-E: eyJlbmMiOiJBMTI4R0NNIiwiYWxnIjoiUlNBLU9BRVAifQ.XmuAS3x5r-0MJuwLDdD4jNC6zjsY7HAFNo2VdvGg6jGcj4hZ4NaJgH20ez313H8An9UJrsUj8ERH0R8UyjQu2UGMUnJ5B1ooXFPla0LQEbN_Em3-IG84YPFcWVmEgcs1Fl2jjlKHVqZp04D21UvtgQ4xyPwQ-QDdTxHqyvSCpcE.ACRnQsNuTh1K_C9R.xpLNZ8Cc9faKoOYhss1CD0A4hG4m0M7-LZQ0fISw7NUHwzQs2AEo9ZXfwOvAj1fCbcE96mbKQo8gr7Oq1a2-piYXM1X5yNMcCxEaYyGinpnf6PGqbdr6zbYZdqyJk0KrxWVhKSQchLJaLGJOts4GlpqujSqJObJQcWWbkJQYKG9K7oKsdtMAKsHIVo5-0BCUbjKVnHJNsYwTsI7xn2Om8zGm4A.nBOuiEDssVFHC_N68tDjVA' -H 'X-SAL-ContentType: e7FDDltrTy+tA2HnLovvGL0LFMwT+KkEptGju5wXVTU=' -H 'X-API-RequestId: ae855627-097b-9bb0-a2dd-92925520bce1' -H 'ApiKey: lstzFDEOhfFNMLikKa0am9mgEKLBl49T' -H 'Origin: https://www.morningstar.com' -H 'Connection: keep-alive' -H 'TE: Trailers'"
req = runcurl.execute(curlstr)
if req.status_code == 200:
fund_category = req.json()['categoryName']
print('Fund category Morningstar: ' + fund_category)
return fund_category
else:
print(fund_isin + ': cannot retrieve fund category from Morningstar!')
return ''
def get_yahoo_fund_symbol(fund_isin, fund_exchange=''):
"""
>EXPERIMENTAL< It does not work very well...
Fund_exchange is typically one of 'AMS', 'LSE', 'GER', 'MIL', 'FRA'. If
omitted, it will prompt for interactive selection.
"""
# curl string to obtain the yahoo symbols given the fund name
curlstr = "curl 'https://query1.finance.yahoo.com/v1/finance/search?q=" + fund_isin + "&lang=en-US®ion=US"esCount=6&newsCount=4&enableFuzzyQuery=false"esQueryId=tss_match_phrase_query&multiQuoteQueryId=multi_quote_single_token_query&newsQueryId=news_cie_vespa&enableCb=true&enableNavLinks=true&enableEnhancedTrivialQuery=true' -H 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:82.0) Gecko/20100101 Firefox/82.0' -H 'Accept: */*' -H 'Accept-Language: en-US,en;q=0.5' --compressed -H 'Referer: https://finance.yahoo.com/screener?.tsrc=fin-srch' -H 'Origin: https://finance.yahoo.com' -H 'Connection: keep-alive' -H 'Cookie: B=bvg32mtd9ql9o&b=3&s=uq; GUC=AQABAgFfO-NgIUIgrgSo; PRF=t%3DVGWL.F%252BVDVA.L%252BVVAL.AS%252BVVAL.L%252BVVL.TO%252BZPRX.DE%252BEURUSD%253DX%252B%255EGSPC%252BES%253DF%252BBRK-B%252BBRKB%252BWORK%252BVEIEX%252BXDEV.DE%252BVVAL.SW%26qct%3DtrendArea; ucs=eup=2; A1=d=AQABBPVPhl4CEJQlhU7jN3yvYRgayeLumysFEgABAgHjO18hYO2Nb2UBACAAAAcIOFWdWrdiwL8&S=AQAAAji0MRIc7MuC98UsHfOepqk; A3=d=AQABBPVPhl4CEJQlhU7jN3yvYRgayeLumysFEgABAgHjO18hYO2Nb2UBACAAAAcIOFWdWrdiwL8&S=AQAAAji0MRIc7MuC98UsHfOepqk; A1S=d=AQABBPVPhl4CEJQlhU7jN3yvYRgayeLumysFEgABAgHjO18hYO2Nb2UBACAAAAcIOFWdWrdiwL8&S=AQAAAji0MRIc7MuC98UsHfOepqk&j=GDPR; thamba=1' -H 'Pragma: no-cache' -H 'Cache-Control: no-cache' -H 'TE: Trailers'"
req = runcurl.execute(curlstr)
# extract the yahoo fund symbol
if req.status_code == 200:
n = len(req.json()['quotes'])
else:
print('Fund not found on Yahoo!')
return ''
# search the exchange when not provided
if not fund_exchange:
# interactive exchange selection
if n > 1:
print('Multiple funds found:')
funds = [(quote['symbol'], quote['shortname'], quote['exchange']) for quote in req.json()['quotes']]
print(*enumerate(funds, 1), sep='\n')
fund_number = input('Please make a selection: ')
fund_exchange = req.json()['quotes'][int(fund_number) - 1]['exchange']
else:
fund_exchange = req.json()['quotes'][0]['exchange']
print('Fund exchange Yahoo: ' + fund_exchange)
# select the symbol corresponding to the desired exchange
fund_symbol = list(filter(lambda quote: quote['exchange'] == fund_exchange, req.json()['quotes']))[0]['symbol']
print('Fund symbol Yahoo: ' + fund_symbol)
return fund_symbol
def get_yahoo_fund_currency(fund_symbol):
try:
json = pd.read_json('https://query1.finance.yahoo.com/v8/finance/chart/' + fund_symbol)
except HTTPError:
print(fund_symbol + ': cannot retrieve fund currency from Yahoo!')
return ''
fund_currency = json['chart']['result'][0]['meta']['currency'].upper()
print(fund_symbol + ': fund currency from Yahoo is ' + fund_currency)
return fund_currency
def get_yahoo_price_data(fund_symbol, cache_dir='price\\', suffixes=None):
if suffixes is None:
suffixes = ['.F', '.DE']
# try to load the price data from file, if not skip to next
try:
price = pd.read_pickle(cache_dir + fund_symbol.upper())
# print the date interval of price data
print(fund_symbol.upper() + ' price data interval is ' + str(price.index[0]) + ' to ' + str(price.index[-1]))
return price, fund_symbol
except FileNotFoundError:
pass
# read the price data
try:
price = pd.read_csv('https://query1.finance.yahoo.com/v7/finance/download/' + fund_symbol +
'?period1=0&period2=10000000000&interval=1d&events=history&includeAdjustedClose=true',
index_col=['Date'])['Adj Close'].rename('Price')
price.index = pd.to_datetime(price.index).to_period("B")
# print the date interval of price data
print(fund_symbol.upper() + ' price data interval is ' + str(price.index[0]) + ' to ' + str(price.index[-1]))
# save the price data to file
price.to_pickle(cache_dir + fund_symbol.upper())
except HTTPError:
print(fund_symbol + ': cannot retrieve fund price data from Yahoo!')
if not suffixes:
return pd.Series(), fund_symbol
else:
# get the next symbol from the available list
fund_symbol = fund_symbol.split('.')[0] + suffixes.pop()
# try reading the same symbol from another exchange
price, _ = get_yahoo_price_data(fund_symbol, suffixes=suffixes)
return price, fund_symbol
def get_av_price_data(fund_symbol, cache_dir='price\\'):
# try to load the price data from file, if not skip to next
try:
return pd.read_pickle(cache_dir + fund_symbol.upper())
except FileNotFoundError:
pass
# check that an alphavantage API key is set in the system environment variables
if not os.getenv('ALPHAVANTAGE_API_KEY'):
raise Exception("Please set 'ALPHAVANTAGE_API_KEY' environment variable!")
# read the price data
try:
price = web.DataReader(fund_symbol.upper(), 'av-daily-adjusted')['adjusted close'].rename('Price')
except ValueError:
print(fund_symbol + ": could not retrieve price data from AlphaVantage. Check valid ticker.")
return pd.Series()
price.index = pd.to_datetime(price.index).to_period("B")
price.index.name = 'Date'
# save the price data to file
price.to_pickle(cache_dir + fund_symbol.upper())
print(fund_symbol.upper() + ' price data interval: ' + str(price.index[0]) + ' to ' + str(price.index[-1]))
return price
def get_av_forex_data(base_currency, to_currency, cache_dir='forex\\'):
# try to load the currency conversion data from file, if not skip
try:
return pd.read_pickle(cache_dir + base_currency + '-' + to_currency)
except FileNotFoundError:
pass
# check that an alphavantage API key is set in the system environment variables
if not os.getenv('ALPHAVANTAGE_API_KEY'):
raise Exception("Please set 'ALPHAVANTAGE_API_KEY' environment variable!")
# read the currency data
try:
fx = web.DataReader(base_currency + '/' + to_currency, 'av-forex-daily')['close'].rename('FX')
except ValueError:
print("Could not retrieve forex data from AlphaVantage. Check valid currency.")
return pd.Series()
fx.index = pd.to_datetime(fx.index).to_period("B")
fx.index.name = 'Date'
# save the currency data to file
fx.to_pickle(cache_dir + base_currency + '-' + to_currency)
return fx
def get_excel_price_data(file):
# read the price data from file
price = pd.read_excel(file, sheet_name='Price_Daily', index_col=0)
# convert the index to date period format
price.index = pd.to_datetime(price.index, format='%Y%m%d').to_period("B")
return price
def get_csv_price_data(file):
# read the price data from file
price = pd.read_csv(file, index_col=0).iloc[:, 0].rename('Price')
# convert the index to date period format
price.index = pd.to_datetime(price.index, format='%Y%m%d').to_period("B")
return price
def get_csv_boe_usd_eur_rate(file='forex\\Bank of England Database.csv'):
# read the EUR/USD exchange data from csv
fx = pd.read_csv(file, index_col=0).iloc[:, 0]
fx.name = 'FX'
# convert the index to daily period format
fx.index = pd.to_datetime(fx.index).to_period('B')
return fx
def calc_usd_eur_return(freq='daily'):
# get the EUR/USD exchange data and 1/x to get USD/EUR rate
fx = get_csv_boe_usd_eur_rate() ** (-1)
# calculate the rate of return
if freq == 'monthly':
rfx = calc_return(fx.sort_index().resample("M").last(), freq='daily')
elif freq == 'daily':
rfx = calc_return(fx.sort_index(), freq='daily')
else:
raise Exception("freq must be either 'daily' or 'monthly'!")
rfx.rename(columns={'Return': 'rFX'}, inplace=True)
return rfx
def get_csv_eonia_rate(freq='daily', file='eonia\\data.csv'):
# read the eonia rate data from csv file
rf = pd.read_csv(file, index_col=0, skiprows=4, parse_dates=True, header=0, names=['Date', 'RF_EUR']) / 100
# convert the index to daily period format
rf.index = pd.to_datetime(rf.index).to_period('B')
if freq == 'monthly':
return rf.resample("M").last()
elif freq == 'daily':
return rf
else:
raise Exception("freq must be either 'daily' or 'monthly'!")
def convert_factor_data_to_eur(factor_data):
if factor_data.empty:
return pd.DataFrame()
# detect whether factor data has daily or monthly resolution
if factor_data.index.dtype == 'period[B]':
freq = 'daily'
elif factor_data.index.dtype == 'period[M]':
freq = 'monthly'
else:
raise Exception('Unable to detect factor data resolution!')
# join the USD/EUR rate data
factor_data = factor_data.join(calc_usd_eur_return(freq=freq), how='inner')
# join the European risk-free rate data
factor_data = factor_data.join(get_csv_eonia_rate(freq=freq), how='inner')
# calculate excess market rete of return in EUR
factor_data_eur = pd.DataFrame()
factor_data_eur['Mkt-RF'] = (1 + factor_data['rFX']) ** (-1) * \
(1 + factor_data['Mkt-RF'] + factor_data['RF']) - 1 - factor_data['RF_EUR']
# add column with EUR risk-free rate of return
factor_data_eur['RF'] = factor_data['RF_EUR']
# convert long-short USD factor data to EUR
if 'SMB' in factor_data.columns:
factor_data_eur['SMB'] = (1 + factor_data['rFX']) ** (-1) * factor_data['SMB']
if 'HML' in factor_data.columns:
factor_data_eur['HML'] = (1 + factor_data['rFX']) ** (-1) * factor_data['HML']
if 'RMW' in factor_data.columns:
factor_data_eur['RMW'] = (1 + factor_data['rFX']) ** (-1) * factor_data['RMW']
if 'CMA' in factor_data.columns:
factor_data_eur['CMA'] = (1 + factor_data['rFX']) ** (-1) * factor_data['CMA']
if 'WML' in factor_data.columns:
factor_data_eur['WML'] = (1 + factor_data['rFX']) ** (-1) * factor_data['WML']
return factor_data_eur
def convert_price_currency(price, fund_currency, to_currency):
fx = get_av_forex_data(base_currency=fund_currency, to_currency=to_currency)
df = pd.merge(price, fx, left_index=True, right_index=True)
return (df.Price * df.FX).rename('Price')
def calc_return(price, freq):
if freq == 'daily':
# calculate daily returns
ret = price.pct_change()[1:].to_frame(name='Return')
return ret[ret.all(1)]
elif freq == 'monthly':
# calculate monthly returns
return price.resample("M").last().pct_change()[1:].to_frame(name='Return')
else:
raise Exception("freq must be either 'daily' or 'monthly'!")
def get_famafrench_data(name_factor_data, name_mom_data, cache_dir='famafrench\\'):
if name_factor_data:
try:
factor_data = pd.read_pickle(cache_dir + name_factor_data)
except FileNotFoundError:
factor_data = web.DataReader(name_factor_data, 'famafrench')[0] / 100
factor_data.to_pickle(cache_dir + name_factor_data)
if name_mom_data:
try:
mom_data = pd.read_pickle(cache_dir + name_mom_data)
except FileNotFoundError:
mom_data = web.DataReader(name_mom_data, 'famafrench')[0] / 100
mom_data.to_pickle(cache_dir + name_mom_data)
if 'WML' not in mom_data.columns:
mom_data = mom_data.iloc[:, 0].to_frame(name='WML')
# merge the momentum with the other factors
factor_data = factor_data.merge(mom_data, left_index=True, right_index=True)
if 'period' not in str(factor_data.index.dtype):
factor_data.index = factor_data.index.to_period("B")
else:
factor_data = pd.DataFrame()
return factor_data
def calc_famafrench_regression(factor_data, fund_data, fund_symbol, quiet=False):
if not factor_data.empty:
x = fund_data.merge(factor_data, left_index=True, right_index=True)
x['Return-RF'] = x['Return'] - x['RF']
y = x['Return-RF']
x = x.drop(['RF', 'Return', 'Return-RF'], axis=1)
x = sm.add_constant(x[['Mkt-RF', 'SMB', 'HML', 'WML']])
model = sm.OLS(y, x).fit(cov_type='HAC', cov_kwds={'maxlags': 1})
model.predict(x)
res = model.params.copy()
res[abs(model.tvalues) < 1.96] = None
res.name = fund_symbol
reg = res.to_frame().transpose()
reg.insert(0, 'R2_adj', model.rsquared_adj)
reg.insert(0, 'N', int(model.nobs))
if not quiet:
print(model.summary())
print('Based on: ' + str(y.index[0]) + ' to ' + str(y.index[-1]))
return reg
else:
return pd.DataFrame()
def get_fund_factor_data(fund_isin, freq):
# obtain the fund category
fund_category = get_morningstar_fund_category(fund_isin)
if freq == 'daily':
# assign the name of daily factor data
if 'US' in fund_category:
name_factor_data = 'F-F_Research_Data_5_Factors_2x3_daily'
name_mom_data = 'F-F_Momentum_Factor_daily'
elif 'Global' in fund_category and 'Emerging' not in fund_category:
name_factor_data = 'Developed_5_Factors_Daily'
name_mom_data = 'Developed_Mom_Factor_Daily'
elif 'Europe' in fund_category or 'Eurozone' in fund_category:
name_factor_data = 'Europe_5_Factors_Daily'
name_mom_data = 'Europe_Mom_Factor_Daily'
else:
name_factor_data = None
name_mom_data = None
elif freq == 'monthly':
if 'Emerging' in fund_category:
name_factor_data = 'Emerging_5_Factors'
name_mom_data = 'Emerging_MOM_Factor'
elif 'US' in fund_category:
name_factor_data = 'F-F_Research_Data_5_Factors_2x3'
name_mom_data = 'F-F_Momentum_Factor'
elif 'Global' in fund_category and 'Emerging' not in fund_category:
name_factor_data = 'Developed_5_Factors'
name_mom_data = 'Developed_Mom_Factor'
elif 'Europe' in fund_category or 'Eurozone' in fund_category:
name_factor_data = 'Europe_5_Factors'
name_mom_data = 'Europe_Mom_Factor'
else:
name_factor_data = None
name_mom_data = None
else:
raise Exception("`freq´ must be one of 'daily' or 'monthly'")
if name_factor_data:
print(freq + ' factor data: ' + name_factor_data)
else:
print(freq + ' factor data not found!')
# retrieve fama-french daily factor data
return get_famafrench_data(name_factor_data, name_mom_data)
def run_fund_regression(fund_symbol, fund_isin, freq, currency, quiet=False):
print('\nRunning ' + freq + ' factor regression in ' + currency + ' for ' + fund_symbol + ' (' + fund_isin + ')')
# check requested currency for regression
if currency not in ['EUR', 'USD']:
raise Exception("currency for regression must be either 'EUR' or 'USD'!")
# retrieve fama-french daily factor data, convert currency if needed
factor_data = (get_fund_factor_data(fund_isin, freq=freq) if currency.upper() == 'USD' else
convert_factor_data_to_eur(get_fund_factor_data(fund_isin, freq=freq)))
if factor_data.empty:
print(fund_isin + ': cannot retrieve factor data for this fund. Aborted.')
return pd.DataFrame()
# get the fund price data and possibly update the symbol
price, fund_symbol = get_yahoo_price_data(fund_symbol)
if price.empty:
return pd.DataFrame()
# get the fund currency
fund_currency = get_yahoo_fund_currency(fund_symbol)
if currency == 'USD':
if fund_currency != 'USD':
# currency conversion of non USD price (if currency == 'USD')
price = convert_price_currency(price, fund_currency, to_currency='USD')
elif currency == 'EUR':
if fund_currency != 'EUR':
# currency conversion of non EUR price (if currency == 'EUR')
price = convert_price_currency(price, fund_currency, to_currency='EUR')
# calculate daily returns
ret = calc_return(price, freq=freq)
# calculating regression
return calc_famafrench_regression(factor_data, ret, fund_symbol, quiet=quiet)
def run_regressions(currency='EUR', freq='daily', fund_info=pd.read_excel('..//Instruments.xlsx', sheet_name='Equity', index_col=0)):
# initialize dataframe for regression results
reg = pd.DataFrame()
for fund in fund_info.itertuples():
if isinstance(fund.Symbol, str):
reg = reg.append(run_fund_regression(fund.Symbol, fund.Index, freq=freq, currency='EUR', quiet=True))
print('\nFactor Regression Results')
print(reg)
suffix = (currency.lower() if currency.upper() == 'EUR' else currency.lower())
reg.to_csv('results\\reg_all_' + freq + '_' + suffix + '.csv', encoding='utf-8', index_label='ISIN')
return reg
def run_regressions_local_data(currency='EUR', freq='daily', prices_dir='nav data\\'):
# get fund price data file
fund_info = pd.DataFrame(glob(prices_dir + '*.csv'), columns=['FilePath'])
# initialize dataframe for regression results
reg = pd.DataFrame()
for fund in fund_info.itertuples():
# extract ISIN from filename
isin = ntpath.splitext(ntpath.basename(fund.FilePath))[0].split('-')[0]
print('\nNow processing ' + get_morningstar_fund_name(isin))
# retrieve fama-french factor data
factor_data = (convert_factor_data_to_eur(get_fund_factor_data(isin, freq=freq))
if currency.upper() == 'EUR' else get_fund_factor_data(isin, freq=freq))
# read the price data
price = get_csv_price_data(fund.FilePath)
# extract the fund currency from the filename
fund_currency = ntpath.splitext(ntpath.basename(fund.FilePath))[0].split('-')[1]
# currency conversion of non EUR price
if fund_currency != 'EUR':
price = convert_price_currency(price, fund_currency, to_currency='EUR')
if not factor_data.empty:
# calculate returns
ret = calc_return(price, freq=freq)
# calculating regression
reg = reg.append(calc_famafrench_regression(factor_data, ret, isin, quiet=True))
print('\nFactor Regression Results')
print(reg)
suffix = ('_eur' if currency.upper() == 'EUR' else '_usd')
# export daily regression results to csv
reg.to_csv('results\\reg_' + freq + suffix + '.csv', encoding='utf-8', index_label='ISIN')
if __name__ == '__main__':
run_regressions_local_data(currency='EUR', freq='monthly')