-
Notifications
You must be signed in to change notification settings - Fork 0
/
Generali.py
2452 lines (2253 loc) · 120 KB
/
Generali.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
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# Import librairies
import streamlit as st
import numpy as np
import pandas as pd
from dateutil import relativedelta
from pandas.tseries.offsets import DateOffset
from pathlib import Path
from io import BytesIO
import requests
import statsmodels.api as sm
from statsmodels.regression.rolling import RollingOLS
import scipy.stats
from scipy.optimize import minimize
from sklearn.metrics import r2_score
from empyrical import (sharpe_ratio, calmar_ratio, omega_ratio, sortino_ratio,
cagr, annual_volatility, tail_ratio,
up_capture, down_capture,
alpha_beta, up_alpha_beta, down_alpha_beta,
value_at_risk, conditional_value_at_risk,
cum_returns_final, cum_returns
)
from quantstats.stats import drawdown_details, to_drawdown_series
import plotly.graph_objects as go
import plotly.express as px
import plotly.figure_factory as ff
# configuration streamlit
st.set_page_config(layout='wide')
# Fonctions
def sharpe(rdt, risk_free_rdt, period='daily'):
if period=='daily':
risk_free_rate = cagr(risk_free_rdt)/252
elif period=='weelky':
risk_free_rate = cagr(risk_free_rdt)/52
elif period=='monthly':
risk_free_rate = cagr(risk_free_rdt)/12
return sharpe_ratio(rdt, risk_free=risk_free_rate, period='daily')
def diff_date(date1, date2):
date_diff = relativedelta.relativedelta(date2, date1)
date = [f'{abs(date_diff.years)} ans ' if abs(date_diff.years) > 0 else str(),
f'{abs(date_diff.months)} mois ' if abs(
date_diff.months) > 0 else str(),
f'{abs(date_diff.days)} jours' if abs(date_diff.days) > 0 else str()]
return ''.join(date)
def human_format(num, round_to=1):
magnitude = 0
while abs(num) >= 1000:
magnitude += 1
num = round(num / 1000.0, round_to)
return '{:.{}f}{}'.format(num, round_to, ['', 'K', 'M', 'B', 'G'][magnitude])
def custom_styling(val):
color = "red" if val < 0 else "black"
return f"color: {color}"
def highlight(s):
if s.Nom == 'TOTAL':
return ['background-color: red'] * len(s)
else:
['background-color: white'] * len(s)
def tracking_error(r_a, r_b):
'''
Returns the tracking error between two return series.
This method is used in Sharpe Analysis minimization problem.
'''
return (((r_a - r_b)**2).sum())**(0.5)
def style_analysis_tracking_error(weights, ref_r, bb_r):
'''
Sharpe style analysis objective function.
Returns the tracking error between the reference returns
and a portfolio of building block returns held with given weights.
'''
return tracking_error(ref_r, (weights*bb_r).sum(axis=1))
def style_analysis(dep_var, exp_vars):
'''
Sharpe style analysis optimization problem.
Returns the optimal weights that minimizes the tracking error between a portfolio
of the explanatory (return) variables and the dependent (return) variable.
'''
# dep_var is expected to be a pd.Series
if isinstance(dep_var, pd.DataFrame):
dep_var = dep_var[dep_var.columns[0]]
n = exp_vars.shape[1]
init_guess = np.repeat(1/n, n)
weights_const = {
'type': 'eq',
'fun': lambda weights: 1 - np.sum(weights)
}
solution = minimize(style_analysis_tracking_error,
init_guess,
method='SLSQP',
options={'disp': False},
args=(dep_var, exp_vars),
constraints=(weights_const,),
bounds=((0.0, 1.0),)*n)
# weights = pd.Series(solution.x, index=exp_vars.columns)
return solution.x.reshape(1, -1)
def rolling_window(a, window_size):
shape = (a.shape[0] - window_size + 1, window_size) + a.shape[1:]
strides = (a.strides[0],) + a.strides
return np.lib.stride_tricks.as_strided(a, shape=shape, strides=strides)
def rolling_style_analysis(dep_var, exp_vars, window_size=52):
data = pd.concat([dep_var, exp_vars], axis=1)
data_index = data.iloc[window_size-1:].index
data_col = exp_vars.columns
nbre_col = exp_vars.shape[1]+1
data = data.to_numpy()
data_rolling = rolling_window(data, window_size=window_size)
weights = np.concatenate([100*style_analysis(data_roll[:, 0],
data_roll[:, 1:nbre_col]) for data_roll in data_rolling])
return pd.DataFrame(weights, columns=data_col, index=data_index)
def roll_cagr(data, window_size=252, period='daily'):
data_index = data.iloc[window_size-1:].index
data_col = data.columns
data = data.to_numpy()
data_rolling = rolling_window(data, window_size=window_size)
rolling_cagr = [cagr(data_roll, period=period) for data_roll in data_rolling]
return pd.DataFrame(rolling_cagr, columns=data_col, index=data_index)
def compound_returns(s, start=100):
'''
Compound a pd.Dataframe or pd.Series of returns from an initial default value equal to 100.
In the former case, the method compounds the returns for every column (Series) by using pd.aggregate.
The method returns a pd.Dataframe or pd.Series - using cumprod().
See also the COMPOUND method.
'''
if isinstance(s, pd.DataFrame):
return s.aggregate(compound_returns, start=start)
elif isinstance(s, pd.Series):
return start * (1 + s).cumprod()
else:
raise TypeError("Expected pd.DataFrame or pd.Series")
def drawdown(rets: pd.Series, start=1000):
'''
Compute the drawdowns of an input pd.Series of returns.
The method returns a dataframe containing:
1. the associated wealth index (for an hypothetical starting investment of $1000)
2. all previous peaks
3. the drawdowns
'''
wealth_index = compound_returns(rets, start=start)
previous_peaks = wealth_index.cummax()
drawdowns = (wealth_index - previous_peaks) / previous_peaks
df = pd.DataFrame(
{"Wealth": wealth_index, "Peaks": previous_peaks, "Drawdown": drawdowns})
return df
def summary_stats_glissant(s, period='daily'):
if period == 'daily':
lag = 252
elif period == 'weekly':
lag = 52
else:
lag = 12
stats = pd.DataFrame()
stats['Perf 2023'] = 100 *s.aggregate(lambda x: cum_returns_final(x['2023']) if x['2023'].isnull().sum() == 0 else np.nan)
stats['Perf 2022'] = 100 *s.aggregate(lambda x: cum_returns_final(x['2022']) if x['2022'].isnull().sum() == 0 else np.nan)
stats['Perf 2021'] = 100 *s.aggregate(lambda x: cum_returns_final(x['2021']) if x['2021'].isnull().sum() == 0 else np.nan)
stats['Perf 1 an'] = 100 *s.aggregate(lambda x: cum_returns_final(x[-lag:-1]) if x[-260:].isnull().sum() == 0 else np.nan)
stats['Perf 3 ans'] = 100 *s.aggregate(lambda x: cum_returns_final(x[-3*lag:-1]) if x[-3*260:].isnull().sum() == 0 else np.nan)
stats['Perf 5 ans'] = 100 *s.aggregate(lambda x: cum_returns_final(x[-5*lag:-1]) if x[-5*260:].isnull().sum() == 0 else np.nan)
stats["Ann. vol 1 an"] = 100*s.aggregate(lambda x: annual_volatility(x.iloc[-lag:-1], period=period) if x[-260:].isnull().sum() == 0 else np.nan)
stats["Ann. vol 3 ans"] = 100 *s.aggregate(lambda x: annual_volatility(x.iloc[-3*lag:-1], period=period) if x[-3*260:].isnull().sum() == 0 else np.nan)
stats["Ann. vol 5 ans"] = 100 *s.aggregate(lambda x: annual_volatility(x.iloc[-5*lag:-1], period=period) if x[-5*260:].isnull().sum() == 0 else np.nan)
stats["Sharpe ratio 1 an"] = s.aggregate(lambda x: sharpe(x[-260:],
risk_free_rdt=bench['Eonia Capitalization Index 7 Day'].iloc[-260:].pct_change(),
period=period) if x[-260:].isnull().sum() == 0 else np.nan
)
stats["Sharpe ratio 3 ans"] = s.aggregate(lambda x: sharpe(x[-3*260:],
risk_free_rdt=bench['Eonia Capitalization Index 7 Day'].iloc[-3*260:].pct_change(),
period=period) if x[-3*260:].isnull().sum() == 0 else np.nan
)
stats["Sharpe ratio 5 ans"] = s.aggregate(lambda x: sharpe(x[-5*260:],
risk_free_rdt=bench['Eonia Capitalization Index 7 Day'].iloc[-5*260:].pct_change(),
period=period) if x[-5*260:].isnull().sum() == 0 else np.nan
)
return stats.T
def summary_stats_perf(s, risk_free_rate=0.00, period='daily'):
if period == 'daily':
lag = 252
elif period == 'weekly':
lag = 52
else:
lag = 12
stats = pd.DataFrame()
stats["Date début"] = s.aggregate(
lambda r: r.index[0].strftime("%d/%m/%Y"))
stats["Date fin"] = s.aggregate(lambda r: r.index[-1].strftime("%d/%m/%Y"))
stats["Taux sans risque"] = 100*risk_free_rate
stats["Ann. return"] = 100*s.aggregate(cagr, period=period)
stats["% Rdt>0"] = s.aggregate(lambda r: 100*r[r > 0].size/r.size)
stats["Average up"] = s.aggregate(lambda r: 100*r[r >= 0].mean())
stats["Average Down"] = s.aggregate(lambda r: 100*r[r < 0].mean())
stats['Up ratio'] = s.aggregate(
lambda r: 100*up_capture(r, df['indice'], period=period))
stats['Down ratio'] = s.aggregate(
lambda r: 100*down_capture(r, df['indice'], period=period))
stats["Skewness"] = s.aggregate(lambda x: x.skew())
stats["kurtosis"] = s.aggregate(
lambda x: x.kurtosis()) # s.aggregate(kurtosis)
stats['Is Normal ?'] = s.aggregate(
lambda x: scipy.stats.jarque_bera(x)[1] >= 0.05)
return stats.T
def summary_stats_risk(s, period='daily', var_level=0.05):
if period == 'daily':
lag = 252
elif period == 'weekly':
lag = 52
else:
lag = 12
stats = pd.DataFrame()
stats["Ann. vol"] = 100 * \
s.aggregate(lambda x: annual_volatility(x, period=period))
stats["Historic Var"] = 100*s.aggregate(value_at_risk, cutoff=var_level)
stats["Historic CVar"] = 100 * \
s.aggregate(conditional_value_at_risk, cutoff=var_level)
stats["Max Drawdown"] = 100 * \
s.aggregate(lambda r: drawdown(r)["Drawdown"].min())
stats["Drawdown average"] = 100*s.aggregate(lambda r: drawdown(
r).loc[drawdown(r)['Drawdown'] < 0, "Drawdown"].mean())
stats['Tail'] = s.aggregate(tail_ratio)
return stats.T
def summary_stats_ratio(s, risk_free_rate=0.00, period='daily'):
if period == 'daily':
lag = 252
elif period == 'weekly':
lag = 52
else:
lag = 12
stats = pd.DataFrame()
stats["Sharpe ratio"] = s.aggregate(
sharpe_ratio, risk_free=risk_free_rate, period=period)
stats["Calmar Ratio"] = s.aggregate(calmar_ratio)
stats["Omega Ratio"] = s.aggregate(
omega_ratio, risk_free=risk_free_rate, required_return=0.0)
stats["Sortino Ratio"] = s.aggregate(sortino_ratio, required_return=0.0)
return stats.T
def calcul_perf_glissant(portef, bench, window=52, period='weekly'):
rdt_portef_glissant = 100*roll_cagr(portef.to_frame(), window_size=window, period=period)
rdt_bench_glissant = 100*roll_cagr(bench.to_frame(), window_size=window, period=period)
data = pd.concat([rdt_portef_glissant, rdt_bench_glissant], axis=1)
ecart_perf_glissant = data.iloc[:,0]-data.iloc[:,1]
return rdt_portef_glissant, rdt_bench_glissant, ecart_perf_glissant
def graph_bar(y, colors, texte='AuM'):
fig = go.Figure([go.Bar(name=col,
x=y.index,
y=y[col],
customdata=np.stack((y[col].apply(lambda x: human_format(x, 0)),
y.sum(1).apply(
lambda x: human_format(x, 0))
),
axis=-1),
xhoverformat="%Y",
xperiodalignment="middle",
hovertemplate='<br>'.join(['Année: %{x}',
texte +
': %{customdata[0]}€',
'<b>Total: %{customdata[1]}€</b>',
]
),
marker_color=colors[col]
)
for col in y.columns]
)
fig.update_layout(barmode='relative',
legend=dict(orientation="h",
yanchor="bottom",
y=1.02,
xanchor="left",
entrywidth=100,
x=0),
uniformtext_mode='hide'
)
fig.update_xaxes(ticklabelmode="period",
tickformat="%Y",
)
fig.update_yaxes(fixedrange=False)
return fig
def graph_bar_Nbre_fonds(y, colors, texte='Nombre'):
fig = go.Figure([go.Bar(name=col,
x=y.index,
y=y[col],
customdata=np.stack((y[col],
y.sum(1)
),
axis=-1),
xhoverformat="%Y",
xperiodalignment="middle",
hovertemplate='<br>'.join(['Année: %{x}',
texte +
': %{customdata[0]}',
'<b>Total: %{customdata[1]}</b>',
]
),
marker_color=colors[col]
)
for col in y.columns]
)
fig.update_layout(barmode='relative',
legend=dict(orientation="h",
yanchor="bottom",
y=1.02,
xanchor="left",
entrywidth=100,
x=0),
uniformtext_mode='hide'
)
fig.update_xaxes(ticklabelmode="period",
tickformat="%Y",
)
fig.update_yaxes(fixedrange=False)
return fig
def selection_fonds(i, key=''):
cols = st.columns(5)
# Choix de l'AM
selected_am = cols[0].selectbox("Choix de la société de gestion",
desc['FUND_MGMT_COMPANY'].sort_values().unique(),
index=i,
key='selected_am'+str(key)+str(i)
)
# Choix du type de fonds
type_fonds = cols[1].multiselect("Nature du fonds",
desc.query(
"FUND_MGMT_COMPANY==@selected_am")['FUND_TYP'].sort_values().unique(),
key='type_fonds'+str(key)+str(i)
)
if not type_fonds:
type_fonds = desc.query(
"FUND_MGMT_COMPANY==@selected_am")['FUND_TYP'].sort_values().unique()
# Choix de la classe d'actifs
classes = ['Toutes classes',
*desc.query("FUND_MGMT_COMPANY==@selected_am & FUND_TYP in @type_fonds")['FUND_ASSET_CLASS_FOCUS'].sort_values().unique()
]
selected_classe = cols[2].selectbox("Choix de la classe d'actif",
classes,
key='selected_classe'+str(key)+str(i)
)
if selected_classe == 'Toutes classes':
fonds_sgp = desc.query(
"FUND_MGMT_COMPANY == @selected_am & FUND_TYP in @type_fonds")['LONG_COMP_NAME']
else:
fonds_sgp = desc.query(
"FUND_ASSET_CLASS_FOCUS==@selected_classe & FUND_MGMT_COMPANY == @selected_am & FUND_TYP in @type_fonds")['LONG_COMP_NAME']
# Choix du fonds
selected_fonds = cols[3].selectbox("Choix du fonds",
vl[fonds_sgp].columns.sort_values(),
key='selected_fonds'+str(key)+str(i),
index=i
)
# Pondérations
weights = cols[4].number_input("Poids:",
min_value=0.0,
max_value=100.0,
value=100/nombre_fonds,
key='poids'+str(key)+str(i),
step=0.5
)
return selected_fonds, weights
def selection_indice(i, key=''):
cols = st.columns(3)
# Choix de l'indice de référence
selected_classe_actif = cols[0].selectbox("Classe d'actifs",
desc_indice['security type'].sort_values(
).unique(),
key='selected_classe_actif' +
str(key)+str(i)
)
selected_indices = cols[1].selectbox("Choix de l'indice",
desc_indice.query("`security type`==@selected_classe_actif").sort_values(
['security type', 'description'])['description'].unique(),
key='selected_indices'+str(key)+str(i)
)
# Pondérations
weights = cols[2].number_input("Poids:",
min_value=0.0,
max_value=100.0,
value=100/nombre_fonds,
key='poids'+str(key)+str(i),
step=0.5
)
return selected_indices, weights
def selection_indice_analyse_style(i, key=''):
cols = st.columns(2)
# Choix de l'indice de référence
selected_classe_actif = cols[0].selectbox("Classe d'actifs",
desc_indice['security type'].sort_values(
).unique(),
key='selected_classe_actif' +
str(key)+str(i)
)
selected_indices = cols[1].selectbox("Choix de l'indice",
desc_indice.query("`security type`==@selected_classe_actif").sort_values(
['security type', 'description'])['description'].unique(),
key='selected_indices'+str(key)+str(i)
)
return selected_indices
def update_data(file):
aum_update = pd.read_excel(file,
sheet_name='aum',
index_col=0,
header=0,
skiprows=[1, 2]).sort_index()
vl_update = pd.read_excel(file,
sheet_name='vl',
index_col=0,
header=0,
skiprows=[1, 2]
).sort_index()
bench_update = pd.read_excel(file,
sheet_name='indices',
index_col=0,
header=0,
skiprows=[1, 2]
).sort_index().dropna(axis=1, how='all').query('~index.duplicated()')
desc_update = pd.read_excel(file,
sheet_name='desc',
index_col=0,
).reset_index()
desc_update['FUND_ASSET_CLASS_FOCUS'] = desc_update['FUND_ASSET_CLASS_FOCUS'].fillna(
'Non renseigné')
desc_update['FUND_MGMT_COMPANY'] = desc_update['FUND_MGMT_COMPANY'].replace(['H2O AM LLP', 'H2O Am Europe SASU', 'Myria Asset Management SAS',
'Myria Asset Management/France', 'Comgest Growth PLC',
'Financiere Arbevel SAS/Fund Parent'],
['H2O AM', 'H2O AM', 'Myria Asset Management',
'Myria Asset Management', 'Comgest SA', 'Financiere Arbevel SAS'],
)
desc_update['FUND_INCEPT_DT'] = pd.to_datetime(
desc_update['FUND_INCEPT_DT'], dayfirst=True)
desc_indice_update = pd.read_excel(file,
sheet_name='desc indice',
index_col=0,
).reset_index()
return aum_update, vl_update, bench_update, desc_update, desc_indice_update
def load_google(code):
url = f"https://drive.google.com/uc?export=download&id={code}"
file = requests.get(url)
bytesio = BytesIO(file.content)
return pd.read_parquet(bytesio)
@st.cache_data
def load_df():
aum = load_google('1-B7Gc12ZcnD-fy_ngS-XQcKyRePERbnw')
vl = load_google('1-JSSHvILfKERBaV-uaGV6PjheBw0OTRZ')
bench = load_google('1-FT7EGsFiN6LiKfSJAkVde5ukfa6NIwZ')
desc = load_google('1-H4arAreH4SioXsLzkkferahTuSzIzcE')
desc_indice = load_google('1-GdUmwFOAA8hLQvn6cU61Etl-x-b4TKZ')
# Gestion des données manquantes
aum = aum.dropna(how='all').dropna(axis=1, how='all')
vl = vl.dropna(how='all').dropna(axis=1, how='all')
bench = bench.dropna(how='all').dropna(axis=1, how='all')
# Nettoyage des fonds en supprimant les fonds non communs entre VL, AUM et desc
fonds = list(set(vl.columns) & set(aum.columns) & set(desc.LONG_COMP_NAME))
aum = aum[fonds].asfreq('B').ffill()
vl = vl[fonds].asfreq('B').ffill()
bench = bench.asfreq('B').ffill()
desc = desc.query("LONG_COMP_NAME in @fonds")
# Construction de la base de données annuelle
vl_annuel = vl.resample('Y').last()
aum_annuel = aum.resample('Y').last()
collecte_annuel = aum_annuel - aum_annuel.shift()*(1+vl_annuel.pct_change())
vl_annuel.columns = pd.MultiIndex.from_tuples(zip(vl_annuel.columns,
vl_annuel.columns.map(desc.set_index('LONG_COMP_NAME')[
'FUND_ASSET_CLASS_FOCUS']),
vl_annuel.columns.map(desc.set_index(
'LONG_COMP_NAME')['FUND_MGMT_COMPANY'])
)
)
aum_annuel.columns = pd.MultiIndex.from_tuples(zip(aum_annuel.columns,
aum_annuel.columns.map(desc.set_index('LONG_COMP_NAME')[
'FUND_ASSET_CLASS_FOCUS']),
aum_annuel.columns.map(desc.set_index(
'LONG_COMP_NAME')['FUND_MGMT_COMPANY']),
)
)
collecte_annuel.columns = pd.MultiIndex.from_tuples(zip(collecte_annuel.columns,
collecte_annuel.columns.map(desc.set_index(
'LONG_COMP_NAME')['FUND_ASSET_CLASS_FOCUS']),
collecte_annuel.columns.map(desc.set_index(
'LONG_COMP_NAME')['FUND_MGMT_COMPANY'])
)
)
# desc.fillna('Non renseigné', inplace=True)
return aum, vl, bench, desc, desc_indice, vl_annuel, aum_annuel, collecte_annuel
# Importation des données
path = Path('/Users/jacques/Library/Mobile Documents/com~apple~CloudDocs/Projets/Analyse fonds/Data')
aum, vl, bench, desc, desc_indice, vl_annuel, aum_annuel, collecte_annuel = load_df()
# Gestion des couleurs vs classes d'actifs
colors = {k: v for k, v in zip(
desc.FUND_ASSET_CLASS_FOCUS.unique(), px.colors.qualitative.Set1)}
# Gestion des différents onglets
tab1, tab2, tab3, tab4, tab5, tab6, tab7, tab8 = st.tabs(["Paramètres",
"Liste des sociétés de gestion",
"Classement des fonds",
"Analyse société de gestion",
"Analyse fonds vs indice",
"Statistiques",
"Analyse de style",
"Update base de données"
]
)
##############################################################################
########################### Onglet Paramètres ################################
##############################################################################
with tab1:
st.header('**Entrez le portefeuille à analyser:**')
nombre_fonds = st.number_input('**Nombre de fonds pour analyse:**',
min_value=1,
step=1,
key='nombre_fonds1'
)
# Portefeuille à analyser
selections = []
for row in range(nombre_fonds):
selections.append(selection_fonds(row))
selection_fonds = pd.DataFrame(selections,
columns=['fonds', 'poids']
)
portef = pd.DataFrame((selection_fonds['poids'].to_numpy()*vl[selection_fonds['fonds']].pct_change().dropna()).sum(1),
columns=['portef']
)
portef = pd.concat([portef,
100*vl[selection_fonds['fonds']].pct_change()],
axis=1)
list_fonds = selection_fonds['fonds'].to_list()
if selection_fonds['poids'].sum() != 100:
st.subheader("**:red[la somme des poids n'est pas égal à 100%]**")
st.header("**Choix de l'indice pour comparaison:**")
indice = st.radio('**Indice ou fonds pour comparaison:**',
('Indice', 'Fonds'),
horizontal=True
)
if indice == 'Indice':
st.subheader("**Entrez l'indice pour comparaison:**")
nombre_fonds = st.number_input('**Nombre de fonds pour analyse:**',
min_value=1,
value=1,
step=1,
key='nombre_fonds2'
)
selections = []
for row in range(nombre_fonds):
selections.append(selection_indice(row, key=1))
bench_indice = pd.DataFrame(selections,
columns=['indice', 'poids']
)
benchmark = pd.DataFrame((bench_indice['poids'].to_numpy()*bench[bench_indice['indice']].pct_change().dropna()).sum(1),
columns=['indice']
)
benchmark = pd.concat([benchmark,
100*bench[bench_indice['indice']].pct_change()
],
axis=1)
list_indice1 = bench_indice['indice'].to_list()
if bench_indice['poids'].sum() != 100:
st.subheader("**:red[la somme des poids n'est pas égal à 100%]**")
else:
st.subheader("**Entrez les fonds pour comparaison:**")
nombre_indice = st.number_input('**Nombre indices pour analyse:**',
min_value=1,
value=1,
step=1,
key='nombre_fonds2'
)
selections = []
for row in range(nombre_indice):
selections.append(selection_fonds(row, key=1))
bench_fonds = pd.DataFrame(selections,
columns=['indice', 'poids']
)
benchmark = pd.DataFrame((bench_fonds['poids'].to_numpy()*vl[bench_fonds['indice']].pct_change().dropna()).sum(1),
columns=['indice']
)
benchmark = pd.concat([benchmark,
vl[bench_fonds['indice']].pct_change()
],
axis=1)
list_indice2 = bench_fonds['indice'].to_list()
if bench_fonds['poids'].sum() != 100:
st.write("**la somme des poids n'est pas égal à 100%**")
# Base de données contenant le fonds, l'indice et ses composants
df = pd.concat([portef,
benchmark],
axis=1).dropna()
df_base100 = cum_returns(df/100,
starting_value=100)
df['ecart'] = df['portef'] - df['indice']
df_base100['ecart'] = 100*df_base100['portef']/df_base100['indice']
df_base100.index = df_base100.index.strftime("%d/%m/%Y")
st.dataframe(df_base100[['portef', 'indice']],
use_container_width=True
)
df_base100.index = pd.to_datetime(df_base100.index, dayfirst=True)
# Image
st.image('Img.gif',
use_column_width=True)
######################################################################################################
##################################### Onglet liste des sociétés ######################################
######################################################################################################
with tab2:
aum_annuel_sdg = aum_annuel.groupby(
level=2, axis=1).sum().iloc[-1].to_frame()
aum_annuel_sdg.columns = ['AuM']
aum_annuel_sdg = aum_annuel_sdg.sort_values('AuM', ascending=False)
nbre_sdg, nbre_fds = aum_annuel_sdg.shape[0], vl.shape[1]
aum_description = pd.concat([desc['FUND_MGMT_COMPANY'].value_counts().to_frame(),
aum_annuel_sdg
],
axis=1).rename({'count': 'Nombre de fonds'}, axis=1)
AuM_total_couvert = aum_description['AuM'].sum()
AuM_total_median = aum_description['AuM'].median()
st.write(f'Nombre de sociétés de gestion: {nbre_sdg}')
st.write(f'Nombre de fonds: {nbre_fds}')
st.write(f'Encours total: {human_format(AuM_total_couvert,1)}€')
col1, col2 = st.columns(2)
with col1:
st.write("**Distribution des AuM par classe d'actifs:**")
st.write('')
st.subheader('')
st.dataframe(pd.concat([aum_annuel.ffill().iloc[-1].groupby(level=1)
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%', 'mean']]
.assign(AuM=lambda x: x['mean']*x['count'])
.rename({'count': 'Nombre de fonds'}, axis=1)
.dropna()
[['Nombre de fonds', 'AuM', '10%', '50%', '90%']],
aum_annuel.ffill().iloc[-1]
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%', 'mean']]
.rename({'count': 'Nombre de fonds'})
.to_frame().T
.rename(index=lambda s: 'Total')
.assign(AuM=lambda x: x['mean']*x['Nombre de fonds']),
])
.assign(**{'AuM en %': lambda x: 2*x['AuM']/x['AuM'].sum()})
[['Nombre de fonds', 'AuM', 'AuM en %', '10%', '50%', '90%']]
.style.format({'Nombre de fonds': "{:.0f}",
'AuM en %': "{:.1%}",
'10%': lambda x: human_format(x, 1)+'€',
'50%': lambda x: human_format(x, 1)+'€',
'90%': lambda x: human_format(x, 1)+'€',
'AuM': lambda x: human_format(x, 1)+'€',
}
),
use_container_width=True
)
with col2:
vl_actif = vl.copy()
vl_actif.columns = pd.MultiIndex.from_tuples(zip(vl_actif.columns,
vl_actif.columns.map(desc.set_index('LONG_COMP_NAME')[
'FUND_ASSET_CLASS_FOCUS']),
)
)
st.write("**Distribution par classe d'actifs:**")
cols = st.columns(2, gap='small')
periode = cols[1].radio("**Période pour calcul de performance:**",
('1Y', '3Y', '5Y', '10Y', '20Y'),
index=2,
label_visibility='collapsed',
horizontal=True)
if periode == '1Y':
lag = 260
elif periode == '3Y':
lag = 3*260
elif periode == '5Y':
lag = 5*260
elif periode == '10Y':
lag = 10*260
else:
lag = 20*260
risk_free_rate = cagr(bench.iloc[-lag:]['Eonia Capitalization Index Capital 5 Day'].pct_change())/252.
analyse = cols[0].radio("**Analyse:**",
('Performances', 'Volatilité', 'Sharpe'),
index=0,
label_visibility='collapsed',
horizontal=True
)
if analyse == 'Performances':
st.dataframe(pd.concat([vl_actif.iloc[-lag:].ffill().apply(lambda x: cagr(x.pct_change()) if x.isnull().sum() == 0 else np.nan)
.groupby(level=[1])
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%']]
.rename({'count': 'Nombre de fonds'}, axis=1)
.dropna(),
vl_actif.iloc[-lag:]
.ffill()
.apply(lambda x: cagr(x.pct_change()) if x.isnull().sum() == 0 else np.nan)
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%']]
.rename({'count': 'Nombre de fonds'}).to_frame().T
.rename(index=lambda s: 'Total')
.dropna()
])
.style.format({'Nombre de fonds': "{:.0f}",
'10%': "{:.2%}",
'50%': "{:.2%}",
'90%': "{:.2%}",
}
),
use_container_width=True
)
elif analyse == 'Volatilité':
st.dataframe(pd.concat([vl_actif.iloc[-lag:].ffill().apply(lambda x: annual_volatility(x.pct_change()) if x.isnull().sum() == 0 else np.nan)
.groupby(level=[1])
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%']]
.rename({'count': 'Nombre de fonds'}, axis=1)
.dropna(),
vl_actif.iloc[-lag:]
.ffill()
.apply(lambda x: annual_volatility(x.pct_change()) if x.isnull().sum() == 0 else np.nan)
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%']]
.rename({'count': 'Nombre de fonds'}).to_frame().T
.rename(index=lambda s: 'Total')
.dropna()
])
.style.format({'Nombre de fonds': "{:.0f}",
'10%': "{:.2%}",
'50%': "{:.2%}",
'90%': "{:.2%}",
}
),
use_container_width=True
)
else:
st.dataframe(pd.concat([vl_actif.iloc[-lag:].ffill().apply(lambda x: sharpe_ratio(x.pct_change(),
risk_free=risk_free_rate)
if x.isnull().sum() == 0 else np.nan)
.groupby(level=[1])
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%']]
.rename({'count': 'Nombre de fonds'}, axis=1)
.dropna(),
vl_actif.iloc[-lag:]
.ffill()
.apply(lambda x: sharpe_ratio(x.pct_change(), risk_free=risk_free_rate)
if x.isnull().sum() == 0 else np.nan)
.describe(percentiles=[0.1, 0.25, 0.5, 0.75, 0.9])[['count', '10%', '50%', '90%']]
.rename({'count': 'Nombre de fonds'}).to_frame().T
.rename(index=lambda s: 'Total')
.dropna()
])
.style.format({'Nombre de fonds': "{:.0f}",
'10%': "{:.2f}",
'50%': "{:.2f}",
'90%': "{:.2f}",
}
),
use_container_width=True
)
col1, col2 = st.columns([1, 1], gap='small')
with col1:
st.caption("AuM global par classes d'actifs")
st.plotly_chart(graph_bar(aum_annuel.groupby(
level=1, axis=1).sum().loc['2005':], texte='AuM', colors=colors))
with col2:
st.caption("Collecte global par classes d'actifs")
st.plotly_chart(graph_bar(collecte_annuel.groupby(
level=1, axis=1).sum().loc['2006':], texte='collecte', colors=colors))
col1, col2 = st.columns([1, 1], gap='small')
with col1:
st.caption('AuM par société de gestion:')
aum_description['AuM moyen par fonds'] = aum_description['AuM'] / \
aum_description['Nombre de fonds']
aum_description = aum_description.sort_values('AuM', ascending=False)
st.dataframe(aum_description.style.format({'AuM': lambda x: human_format(x, 1)+'€',
'AuM moyen par fonds': lambda x: human_format(x, 1)+'€'}
),
use_container_width=True,
height=1250
)
with col2:
col21, col22 = st.columns(2)
with col21:
an = st.number_input('Période en année pour calcul de la collecte:',
min_value=1,
max_value=10,
step=1,
value=3)
with col22:
classes = ['Toutes classes',
*desc['FUND_ASSET_CLASS_FOCUS'].sort_values().unique()
]
selected_classe_collecte = st.selectbox("Choix de la classe d'actif",
classes,
key='selected_classe_collecte2'
)
if selected_classe_collecte == 'Toutes classes':
collecte_annuel_classe = collecte_annuel.droplevel(level=1, axis=1)
else:
collecte_annuel_classe = collecte_annuel.xs(
selected_classe_collecte, axis=1, level=1)
AM_Best_Lower_collecte = pd.concat([collecte_annuel_classe.rolling(an).sum().groupby(level=1, axis=1).sum().iloc[-1].nlargest(10).reset_index(),
collecte_annuel_classe.rolling(an).sum().groupby(level=1, axis=1).sum().iloc[-1].nsmallest(10).reset_index()],
axis=1)
AM_Best_Lower_collecte.columns = [
'Société de gestion ', 'Top 10 collecte', 'Société de gestion', 'Bottom 10 collecte']
AM_Best_Lower_collecte[['Top 10 collecte', 'Bottom 10 collecte']] = AM_Best_Lower_collecte[[
'Top 10 collecte', 'Bottom 10 collecte']].applymap(lambda x: human_format(x, 1)+'€')
Fds_Best_Lower_collecte = pd.concat([collecte_annuel_classe.rolling(an).sum().groupby(level=0, axis=1).sum().iloc[-1].nlargest(10).reset_index(),
collecte_annuel_classe.rolling(an).sum().groupby(level=0, axis=1).sum().iloc[-1].nsmallest(10).reset_index()],
axis=1)
Fds_Best_Lower_collecte.columns = [
'Fonds ', 'Top 10 collecte', 'Fonds', 'Bottom 10 collecte']
Fds_Best_Lower_collecte[['Top 10 collecte', 'Bottom 10 collecte']] = Fds_Best_Lower_collecte[[
'Top 10 collecte', 'Bottom 10 collecte']].applymap(lambda x: human_format(x, 1)+'€')
hide_table_row_index = """
<style>
thead tr th:first-child {display:none}
tbody th {display:none}
</style>
"""
st.markdown(hide_table_row_index, unsafe_allow_html=True)
st.caption('Société de gestion Top/Bottom 10 collecte')
st.table(AM_Best_Lower_collecte)
st.caption('Fonds Top/Bottom 10 collecte')
st.table(Fds_Best_Lower_collecte)
# Nombre de fonds crées par an et par classe d'actifs
col1, col2 = st.columns(2)
with col1:
st.caption("Nombre de fonds crée par an et par classe d'actifs")
fonds_cree = desc.assign(YEAR=desc['FUND_INCEPT_DT'].dt.year)[['YEAR', 'FUND_ASSET_CLASS_FOCUS']]
table = pd.crosstab(fonds_cree['YEAR'], fonds_cree['FUND_ASSET_CLASS_FOCUS']).loc['1990':]
fig = graph_bar_Nbre_fonds(table, colors=colors)
st.plotly_chart(fig,
use_container_width=True)
with col2:
cols = st.columns(2)
st.caption("Création par société de gestion")
lag = cols[0].number_input("Nombre de fonds crée au cours des dernières années:",
min_value=1,
step=1,
value=3)
typ = cols[1].multiselect("Choix type de fonds",
desc.FUND_TYP.unique()
)
if len(typ)>0:
fonds_cree = desc.query("FUND_TYP==@typ")
else:
fonds_cree = desc
fonds_cree = fonds_cree.assign(YEAR=desc['FUND_INCEPT_DT'].dt.year)[['YEAR', 'FUND_MGMT_COMPANY', 'FUND_ASSET_CLASS_FOCUS']]
table = (pd.pivot_table(fonds_cree,
index='YEAR',
columns='FUND_MGMT_COMPANY',
aggfunc='count', fill_value=0)
.apply(lambda x: x.iloc[-lag:].sum())
.sort_values(ascending=False)
.droplevel(0)
.rename_axis(["Société de gestion"])
.rename({0:'Nombre de fonds'})
.to_frame()
.rename({0:'Nombre de fonds crée'}, axis=1)
)
st.dataframe(table[table['Nombre de fonds crée']!=0],
use_container_width=True)
#######################################################################################
################################# classement des fonds ############################
#######################################################################################
with tab3:
st.subheader("Classement des fonds")
cols = st.columns(4)
selection_classe_actif = cols[0].selectbox("Choix de la classe d'actif",
desc['FUND_ASSET_CLASS_FOCUS'].sort_values(
).unique(),
index=2
)
desc2 = desc.query(
"FUND_ASSET_CLASS_FOCUS==@selection_classe_actif").fillna('Non renseigné')
selection_geo = cols[1].multiselect("Zone géographique:",
desc2['FUND_GEO_FOCUS'].sort_values(
).unique(),
)
if len(selection_geo) != 0:
desc2 = desc2.query("FUND_GEO_FOCUS in @selection_geo")
selection_strategy = cols[2].selectbox("Style de gestion:",
['Tout',
*desc2['FUND_STRATEGY'].sort_values().unique()],
index=0
)
if selection_strategy != 'Tout':
desc2 = desc2.query("FUND_STRATEGY in @selection_strategy")
selection_mkt_cap = cols[3].selectbox("Capitalisation boursière:",
['Tout',
*desc2['FUND_MKT_CAP_FOCUS'].sort_values().unique()],
index=0
)
if selection_mkt_cap != 'Tout':
desc2 = desc2.query("FUND_MKT_CAP_FOCUS==@selection_mkt_cap")
perf_comparaison = vl[desc2.LONG_COMP_NAME].agg([lambda x: 100*cagr(x[-260:].pct_change()) if x[-260:].isnull().sum() == 0 else np.nan,
lambda x: 100*cagr(x[-3*260:].pct_change()) if x[-3*260:].isnull().sum() == 0 else np.nan,
lambda x: 100*cagr(x[-5*260:].pct_change()) if x[-5 *260:].isnull().sum() == 0 else np.nan,
lambda x: 100 *cagr(x[-10*260:].pct_change()) if x[-10*260:].isnull().sum() == 0 else np.nan,
lambda x: 100 *annual_volatility(x[-1*260:].pct_change()) if x[-260:].isnull().sum() == 0 else np.nan,
lambda x: 100 *annual_volatility(x[-3*260:].pct_change()) if x[-3*260:].isnull().sum() == 0 else np.nan,
lambda x: 100 *annual_volatility(x[-5*260:].pct_change()) if x[-5*260:].isnull().sum() == 0 else np.nan,
lambda x: 100 *annual_volatility(x[-10*260:].pct_change()) if x[-10*260:].isnull().sum() == 0 else np.nan,
lambda x: sharpe(x[-260:].pct_change(),
risk_free_rdt=bench['Eonia Capitalization Index 7 Day'].iloc[-260:].pct_change())
if x[-260:].isnull().sum() == 0 else np.nan,
lambda x: sharpe(x[-3*260:].pct_change(),
risk_free_rdt=bench['Eonia Capitalization Index 7 Day'].iloc[-3*260:].pct_change()) if x[-3*260:].isnull().sum() == 0 else np.nan,
lambda x: sharpe(x[-5*260:].pct_change(),
risk_free_rdt=bench['Eonia Capitalization Index 7 Day'].iloc[-5*260:].pct_change()) if x[-5*260:].isnull().sum() == 0 else np.nan,
lambda x: sharpe(x[-10*260:].pct_change(),
risk_free_rdt=bench['Eonia Capitalization Index 7 Day'].iloc[-10*260:].pct_change()) if x[-10*260:].isnull().sum() == 0 else np.nan,
],).set_axis(['Perf 1 an', 'Perf 3 ans', 'Perf 5 ans', 'Perf 10 ans',
'Volatilité 1 an', 'Volatilité 3 ans', 'Volatilité 5 ans','Volatilité 10 ans',
'Sharpe 1 an', 'Sharpe 3 ans', 'Sharpe 5 ans', 'Sharpe 10 ans']).T