-
-
Notifications
You must be signed in to change notification settings - Fork 37
/
sss_results_performance.py
197 lines (170 loc) · 10.2 KB
/
sss_results_performance.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
#############################################################################
#
# Version 0.2.5 - Author: Asaf Ravid <[email protected]>
#
# Stock Screener and Scanner - based on yfinance
# Copyright (C) 2021 Asaf Ravid
#
# 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 os
from glob import glob
import csv
import yfinance as yf
import datetime
import numpy
import math
import pandas as pd
import sss
END_DATE_STR = '20210720'
RESULTS_LEN = 35
TASE_MODE = 1
RESULTS_INPUT_FOLDER = "Results/Tase"
YF_DEBUG_MODE = False
results_input_paths = glob(RESULTS_INPUT_FOLDER+'/*/')
# Read Results Input:
def read_engine_results(path, results_filename, max_results, sss_value_names_list, optional_rename):
engine_results_list = []
effective_row_index = 0
sss_value_index = -1
try:
if optional_rename is not None:
os.rename(path+'/'+results_filename, path+'/'+optional_rename)
results_filename = optional_rename
filename_path = path+'/'+results_filename
if os.path.isfile(filename_path):
with open(filename_path, mode='r', newline='') as engine:
reader = csv.reader(engine, delimiter=',')
found_title_row = False
for row in reader:
if not found_title_row:
if row[0] == 'Ticker' or row[0] == 'Symbol':
found_title_row = True
sss_value_index = -1
for sss_value_name in sss_value_names_list:
if sss_value_name in row: sss_value_index = row.index(sss_value_name)
continue
else:
if (sss_value_index >= 0 and float(row[sss_value_index]) > 0.0) or sss_value_index < 0:
row_symbol = row[0]
if 'TLV:' in row_symbol:
row_symbol = row_symbol.replace('TLV:', '')
row_symbol += '.TA'
engine_results_list.append(row_symbol)
effective_row_index += 1
if effective_row_index >= max_results:
break
except Exception as e:
print('Exception {} in [read_engine_results]: path {}, max_results {}, sss_value_name {}'.format(e, path, max_results, sss_value_names_list))
pass
return engine_results_list
def find_start_date_value(symbol_to_check, start_date, pd_database_close_data):
found_date = False
current_date_forward = start_date
current_date_backward = start_date
while not found_date:
current_date_forward_str = current_date_forward.strftime( '%Y-%m-%d')
current_date_backward_str = current_date_backward.strftime('%Y-%m-%d')
try:
start_date_value_check = pd_database_close_data[symbol_to_check].loc[current_date_forward_str]
found_date = True
except Exception as e:
current_date_forward = current_date_forward + datetime.timedelta(days=1)
try:
start_date_value_check = pd_database_close_data[symbol_to_check].loc[current_date_backward_str]
found_date = True
except Exception as e:
current_date_backward = current_date_backward - datetime.timedelta(days=1)
return start_date_value_check
symbol_close_values_db = {}
results_filenames_list = ['sss_engine.csv', 'sss_engine_normalized.csv', 'results_sss.csv', 'results_sss_normalized.csv', 'results_sss_aggregated.csv'] # 'rec_sss.csv', 'rec_sss_1.csv', 'rec_sss_2.csv', 'rec_sss_3.csv', 'rec_sss_4.csv', 'rec_sss_5.csv' ]
optional_rename_list = [None, None, None, None, None ] # 'sss_results.csv', 'sss_results_1.csv', 'sss_results_2.csv', 'sss_results_3.csv', 'sss_results_4.csv', 'sss_results_5.csv']
pd_database_close = None
for results_input_path in results_input_paths:
results_input_path = results_input_path.replace("\\",'/')[:-1]
print('\n Analyzing {}:'.format(results_input_path))
results_date = results_input_path.replace("Results","").replace("Nsr","").replace("All","").replace('Six','').replace("Tase","").replace("/","")[:8]
# year month day
start = datetime.datetime(int(results_date[0:4]), int(results_date[4:6]), int(results_date[6:8]))
end = datetime.datetime(int(END_DATE_STR[0:4]), int(END_DATE_STR[4:6]), int(END_DATE_STR[6:8]))
start = start - datetime.timedelta(days=1)
results_lists = []
gains_lists = []
performance_list = []
performance_indices_list = []
existence_in_db_ratios_list = []
# Comparison Indices:
if pd_database_close is None:
comparison_indices_list = ['SPY', 'QQQ', 'VTWO']
data_start_end_indices_list = yf.download(comparison_indices_list, start=start, end=end, threads=False)
data_start_end_indices_list = data_start_end_indices_list.Close
pd_database_close = data_start_end_indices_list
for index, results_filename in enumerate(results_filenames_list):
results_list = read_engine_results(results_input_path, results_filename, RESULTS_LEN, ['sss_value','value'], optional_rename_list[index])
results_lists.append(results_list)
existing_symbols_list = []
new_symbols_list = []
gains_list = []
performance = None
existence_in_db_ratio = None
if len(results_list):
# Assuming folders are traversed alpphabetically, they will be scanned from oldest to newest, so dates may already exist:
for symbol in results_list:
if symbol in symbol_close_values_db:
existing_symbols_list.append(symbol)
else:
new_symbols_list.append(symbol)
if len(new_symbols_list):
length_was_1 = False
if len(new_symbols_list) == 1: # take 1 from the existing, to get a proper-columned dataframe
new_symbols_list.append(existing_symbols_list[0])
length_was_1 = True
if YF_DEBUG_MODE:
data_start_end_new_symbols_list = None
for i,k in zip(new_symbols_list[0::2], new_symbols_list[1::2]):
data_start_end_new_symbols_list_pair = yf.download([i,k], start=start, end=end, threads=False)
data_start_end_new_symbols_list_pair = data_start_end_new_symbols_list_pair.Close
if data_start_end_new_symbols_list is None:
data_start_end_new_symbols_list = data_start_end_new_symbols_list_pair
else:
data_start_end_new_symbols_list = pd.concat([data_start_end_new_symbols_list, data_start_end_new_symbols_list_pair], axis=1, join="outer")
else:
data_start_end_new_symbols_list = yf.download(new_symbols_list, start=start, end=end, threads=False)
data_start_end_new_symbols_list = data_start_end_new_symbols_list.Close
if length_was_1:
del data_start_end_new_symbols_list[existing_symbols_list[0]]
if pd_database_close is None:
pd_database_close = data_start_end_new_symbols_list
else:
pd_database_close = pd.concat([pd_database_close, data_start_end_new_symbols_list], axis=1, join="outer")
for symbol in results_list:
if symbol not in symbol_close_values_db:
symbol_close_values_db[symbol] = pd_database_close[symbol]
start_date_value = find_start_date_value(symbol, start, pd_database_close)
end_date_value = pd_database_close[symbol][-1]
if not math.isnan(start_date_value) and not math.isnan(end_date_value):
gains_list.append(round(end_date_value/start_date_value-1.0, sss.NUM_ROUND_DECIMALS))
if not len(performance_indices_list):
for comparison_indice in comparison_indices_list:
start_date_value = find_start_date_value(comparison_indice, start, pd_database_close)
end_date_value = pd_database_close[comparison_indice][-1]
if not math.isnan(start_date_value) and not math.isnan(end_date_value):
performance_indices_list.append(round(100.0*(end_date_value / start_date_value - 1.0), sss.NUM_ROUND_DECIMALS))
performance = round(100*numpy.mean(gains_list), sss.NUM_ROUND_DECIMALS)
existence_in_db_ratio = round(float(len(existing_symbols_list))/float(len(results_list)),sss.NUM_ROUND_DECIMALS)
gains_lists.append(gains_list)
performance_list.append(performance)
existence_in_db_ratios_list.append(existence_in_db_ratio)
print(' Performance % between {} and {} of {} is {}. DB existence ({}). {} Indices Performance Comparison: {}'.format(start,end, results_filenames_list, performance_list, existence_in_db_ratios_list, comparison_indices_list, performance_indices_list))