feat: 13장 가치주 포트폴리오 계산 추가( PER,PBR이 낮은 회사 20개)

example/10-calculate-stock-value.py

@@ -1,29 +1,14 @@
 # 패키지 불러오기
-import os
-from urllib.parse import quote_plus
-
-import pymysql
-from sqlalchemy import create_engine
-import pandas as pd
 import numpy as np
-from dotenv import load_dotenv
+import pandas as pd
 
-load_dotenv()
-user = os.getenv('DB_USER')
-pw = os.getenv('DB_PW')
-engine_for_pw = quote_plus(pw)
-host = os.getenv('DB_HOST')
-port = int(os.getenv('DB_PORT'))
-db = os.getenv('DB_DB')
+import quantcommon
 
 # DB 연결
-engine = create_engine(f'mysql+pymysql://{user}:{engine_for_pw}@{host}:{port}/{db}')
-con = pymysql.connect(user=user,
-                      passwd=pw,
-                      host=host,
-                      port=port,
-                      db=db,
-                      charset='utf8')
+common = quantcommon.QuantCommon()
+engine = common.create_engine()
+con = common.connect()
+
 mycursor = con.cursor()
 # 가치 지표 계산
 # 분기 재무제표 불러오기

example/10-financial-statements-one.py

@@ -0,0 +1,114 @@
+import re
+import time
+
+import pandas as pd
+import requests as rq
+from bs4 import BeautifulSoup
+from tqdm import tqdm
+
+import quantcommon
+
+# DB 연결
+common = quantcommon.QuantCommon()
+engine = common.create_engine()
+con = common.connect()
+
+mycursor = con.cursor()
+
+# 재무제표 크롤링
+# 티커리스트 불러오기
+ticker_list = pd.read_sql("""
+select * from kor_ticker
+where 기준일 = (select max(기준일) from kor_ticker) 
+	and 종목구분 = '보통주';
+""", con=engine)
+
+# DB 저장 쿼리
+query = """
+    insert into kor_fs (계정, 기준일, 값, 종목코드, 공시구분)
+    values (%s,%s,%s,%s,%s) as new
+    on duplicate key update
+    값=new.값
+"""
+
+# 오류 발생시 저장할 리스트 생성
+error_list = []
+
+
+# 재무제표 클렌징 함수
+def clean_fs(df, ticker, frequency):
+
+    df = df[~df.loc[:, ~df.columns.isin(['계정'])].isna().all(axis=1)]
+    df = df.drop_duplicates(['계정'], keep='first')
+    df = pd.melt(df, id_vars='계정', var_name='기준일', value_name='값')
+    df = df[~pd.isnull(df['값'])]
+    df['계정'] = df['계정'].replace({'계산에 참여한 계정 펼치기': ''}, regex=True)
+    print(df)
+    df['기준일'] = pd.to_datetime(df['기준일'],
+                               format='%Y/%m') + pd.tseries.offsets.MonthEnd()
+    df['종목코드'] = ticker
+    df['공시구분'] = frequency
+
+    return df
+
+
+i = 0
+ticker = ticker_list['종목코드'][i]
+
+try:
+    # url 생성
+    url = f'https://comp.fnguide.com/SVO2/ASP/SVD_Finance.asp?pGB=1&gicode=A{ticker}'
+
+    # 데이터 받아오기
+    data = pd.read_html(url, displayed_only=False)
+
+    # print([item.head(3) for item in data])
+    # 연간 데이터
+    data_fs_y = pd.concat([
+        data[0].iloc[:, ~data[0].columns.str.contains('전년동기')], data[2],
+        data[4]
+    ])
+    data_fs_y = data_fs_y.rename(columns={data_fs_y.columns[0]: "계정"})
+
+    # 결산년 찾기
+    page_data = rq.get(url)
+    page_data_html = BeautifulSoup(page_data.content, 'html.parser')
+
+    fiscal_data = page_data_html.select('div.corp_group1 > h2')
+    fiscal_data_text = fiscal_data[1].text
+    fiscal_data_text = re.findall('[0-9]+', fiscal_data_text)
+
+    # 결산년에 해당하는 계정만 남기기
+    data_fs_y = data_fs_y.loc[:, (data_fs_y.columns == '계정') | (
+        data_fs_y.columns.str[-2:].isin(fiscal_data_text))]
+
+    # 클렌징
+    data_fs_y_clean = clean_fs(data_fs_y, ticker, 'y')
+
+    # 분기 데이터
+    data_fs_q = pd.concat([
+        data[1].iloc[:, ~data[1].columns.str.contains('전년동기')], data[3],
+        data[5]
+    ])
+    data_fs_q = data_fs_q.rename(columns={data_fs_q.columns[0]: "계정"})
+
+    data_fs_q_clean = clean_fs(data_fs_q, ticker, 'q')
+
+    # 두개 합치기
+    data_fs_bind = pd.concat([data_fs_y_clean, data_fs_q_clean])
+    print(data_fs_bind.head(3))
+
+    # 재무제표 데이터를 DB에 저장
+    args = data_fs_bind.values.tolist()
+    mycursor.executemany(query, args)
+    con.commit()
+
+except Exception as e:
+
+    # 오류 발생시 해당 종목명을 저장하고 다음 루프로 이동
+    print(e)
+    error_list.append(ticker)
+
+# DB 연결 종료
+engine.dispose()
+con.close()

example/10-financial-statements.py

@@ -1,35 +1,21 @@
-import os
-from urllib.parse import quote_plus
+import re
+import time
 
-import pymysql
-from sqlalchemy import create_engine
 import pandas as pd
 import requests as rq
 from bs4 import BeautifulSoup
-import re
 from tqdm import tqdm
-import time
-from dotenv import load_dotenv
 
-load_dotenv()
-user = os.getenv('DB_USER')
-pw = os.getenv('DB_PW')
-engine_for_pw = quote_plus(pw)
-host = os.getenv('DB_HOST')
-port = int(os.getenv('DB_PORT'))
-db = os.getenv('DB_DB')
+import quantcommon
 
 # DB 연결
-engine = create_engine(f'mysql+pymysql://{user}:{engine_for_pw}@{host}:{port}/{db}')
-con = pymysql.connect(user=user,
-                      passwd=pw,
-                      host=host,
-                      port=port,
-                      db=db,
-                      charset='utf8')
+common = quantcommon.QuantCommon()
+engine = common.create_engine()
+con = common.connect()
+
 mycursor = con.cursor()
 
-# 제무재표 크롤링
+# 재무제표 크롤링
 # 티커리스트 불러오기
 ticker_list = pd.read_sql("""
 select * from kor_ticker
@@ -58,7 +44,7 @@ def clean_fs(df, ticker, frequency):
     df = df[~pd.isnull(df['값'])]
     df['계정'] = df['계정'].replace({'계산에 참여한 계정 펼치기': ''}, regex=True)
     df['기준일'] = pd.to_datetime(df['기준일'],
-                               format='%Y-%m') + pd.tseries.offsets.MonthEnd()
+                               format='%Y/%m') + pd.tseries.offsets.MonthEnd()
     df['종목코드'] = ticker
     df['공시구분'] = frequency
 
@@ -75,7 +61,7 @@ for i in tqdm(range(0, len(ticker_list))):
     try:
 
         # url 생성
-        url = f'http://comp.fnguide.com/SVO2/ASP/SVD_Finance.asp?pGB=1&gicode=A{ticker}'
+        url = f'https://comp.fnguide.com/SVO2/ASP/SVD_Finance.asp?pGB=1&gicode=A{ticker}'
 
         # 데이터 받아오기
         data = pd.read_html(url, displayed_only=False)

example/10-stock-data.py

@@ -1,5 +1,3 @@
-import logging
-import os
 import re
 import time
 from io import BytesIO
@@ -9,8 +7,8 @@ import pandas as pd
 import requests as rq
 from tqdm import tqdm
 from bs4 import BeautifulSoup
-import pymysql
 from dotenv import load_dotenv
+import quantcommon
 
 load_dotenv()
 
@@ -125,12 +123,7 @@ def process_for_total_stock(biz_day):
 
 
 def save_ticker(ticker):
-    con = pymysql.connect(user=os.getenv('DB_USER'),
-                          passwd=os.getenv('DB_PW'),
-                          host=os.getenv('DB_HOST'),
-                          port=int(os.getenv('DB_PORT')),
-                          db=os.getenv('DB_DB'),
-                          charset='utf8')
+    con = quantcommon.QuantCommon().connect()
 
     mycursor = con.cursor()
     query = f"""
@@ -175,12 +168,7 @@ def process_for_wics(biz_day):
 
 
 def save_sector(sector):
-    con = pymysql.connect(user=os.getenv('DB_USER'),
-                          passwd=os.getenv('DB_PW'),
-                          host=os.getenv('DB_HOST'),
-                          port=int(os.getenv('DB_PORT')),
-                          db=os.getenv('DB_DB'),
-                          charset='utf8')
+    con = quantcommon.QuantCommon().connect()
 
     mycursor = con.cursor()
     query = f"""
@@ -201,4 +189,4 @@ def save_sector(sector):
 if __name__ == '__main__':
     latest_biz_day = get_latest_biz_day()
     process_for_total_stock(latest_biz_day)
-    # process_for_wics(latest_biz_day)
+    process_for_wics(latest_biz_day)

example/10-stock-price.py

@@ -1,34 +1,21 @@
 # 패키지 불러오기
-import os
-from urllib.parse import quote_plus
 
-import pymysql
-from sqlalchemy import create_engine
-import pandas as pd
-from datetime import date
-from dateutil.relativedelta import relativedelta
-import requests as rq
 import time
-from tqdm import tqdm
+from datetime import date
 from io import BytesIO
-from dotenv import load_dotenv
 
-load_dotenv()
-user = os.getenv('DB_USER')
-pw = os.getenv('DB_PW')
-engine_for_pw = quote_plus(pw)
-host = os.getenv('DB_HOST')
-port = int(os.getenv('DB_PORT'))
-db = os.getenv('DB_DB')
+import pandas as pd
+import requests as rq
+from dateutil.relativedelta import relativedelta
+from tqdm import tqdm
+
+import quantcommon
 
 # DB 연결
-engine = create_engine(f'mysql+pymysql://{user}:{engine_for_pw}@{host}:{port}/{db}')
-con = pymysql.connect(user=user,
-                      passwd=pw,
-                      host=host,
-                      port=port,
-                      db=db,
-                      charset='utf8')
+common = quantcommon.QuantCommon()
+engine = common.create_engine()
+con = common.connect()
+
 mycursor = con.cursor()
 
 # 티커리스트 불러오기

example/13-kor-value-portfolio.py

@@ -0,0 +1,31 @@
+import pandas as pd
+import numpy as np
+import quantcommon
+
+
+#가치주 포트폴리오. PER, PBR이 낮은 회사 20개
+# DB 연결
+engine = quantcommon.QuantCommon().create_engine()
+
+ticker_list = pd.read_sql("""
+select * from kor_ticker
+where 기준일 = (select max(기준일) from kor_ticker) 
+	and 종목구분 = '보통주';
+""", con=engine)
+
+value_list = pd.read_sql("""
+select * from kor_value
+where 기준일 = (select max(기준일) from kor_value);
+""", con=engine)
+
+engine.dispose()
+
+value_list.loc[value_list['값'] <= 0, '값'] = np.nan
+value_pivot = value_list.pivot(index='종목코드', columns='지표', values='값')
+data_bind = ticker_list[['종목코드', '종목명']].merge(value_pivot,
+                                               how='left',
+                                               on='종목코드')
+# print(data_bind.head())
+value_rank = data_bind[['PER', 'PBR']].rank(axis = 0)
+value_sum = value_rank.sum(axis = 1, skipna = False).rank()
+print(data_bind.loc[value_sum <= 20, ['종목코드', '종목명', 'PER', 'PBR']])

example/quantcommon.py

@@ -0,0 +1,28 @@
+import os
+from urllib.parse import quote_plus
+
+import pymysql
+from dotenv import load_dotenv
+from sqlalchemy import create_engine
+
+
+class QuantCommon:
+    def __init__(self):
+        load_dotenv()
+        self.user = os.getenv('DB_USER')
+        self.pw = os.getenv('DB_PW')
+        self.engine_for_pw = quote_plus(self.pw)
+        self.host = os.getenv('DB_HOST')
+        self.port = int(os.getenv('DB_PORT'))
+        self.db = os.getenv('DB_DB')
+
+    def create_engine(self):
+        return create_engine(f'mysql+pymysql://{self.user}:{self.engine_for_pw}@{self.host}:{self.port}/{self.db}')
+
+    def connect(self):
+        return pymysql.connect(user=self.user,
+                      passwd=self.pw,
+                      host=self.host,
+                      port=self.port,
+                      db=self.db,
+                      charset='utf8')