def classify_sectors_uptrend(self,
table_name=None,
time_from=None,
time_to=None,
stocks=None,
from_db=False,
is_industries=False,
separator=""):
# print(self.trend_data)
if stocks is None:
stocks = self.get_trend_slice(table_name=table_name,
time_from=time_from,
time_to=time_to,
from_db=from_db)
# technical, healthcare, financials etc.
# print(stocks)
if stocks is not None:
stocks = Utils.add_first_last_perc_diff(stocks)
stocks[
"industry"] = stocks["sector"] + separator + stocks["industry"]
group_by = "sector" if is_industries is False else "industry"
stocks = stocks.groupby(by=group_by).mean()
return stocks
else:
return stocks
def set_spy_sectors_trend(self):
table_name = TableName.DAY
if self.trend_data is None:
self.classify_sectors_uptrend(table_name=table_name,
time_from="-14d",
time_to="0")
df1 = self.trend_data.iloc[0:int(round(len(self.trend_data) / 2, 0))]
df2 = self.trend_data.iloc[int(round(len(self.trend_data) /
2, 0)):len(self.trend_data)]
# ---------------------------set sectors trend --------------------
self.sectors_trend.append(self.classify_sectors_uptrend(stocks=df1))
self.sectors_trend.append(self.classify_sectors_uptrend(stocks=df2))
# ----------------------------set spy trend-----------------------------
dfs1 = df1[df1.sym == "SPY"]
dfs2 = df2[df2.sym == "SPY"]
self.spy_trend.append(Utils.add_first_last_perc_diff(dfs1))
self.spy_trend.append(Utils.add_first_last_perc_diff(dfs2))
print("set_spy_sectors_trend() - done")
def sectors_uptrend_by_month(self, yf=2017, yt=None, show_chart=True):
# technical, healthcare, financials etc.
sectors = self.db.get_sectors()
sectors_month_stats = pd.DataFrame()
sec_year_stats = {}
# print(sectors)
# self.sectors = [sec]
stocks = sdf.retype(self.db.load_data(TableName.DAY))
stocks.index = pd.to_datetime(stocks.index, utc=True)
stocks['month'] = stocks.index.to_series().dt.month
stocks['year'] = stocks.index.to_series().dt.year
if not yt:
max_year = stocks.year.max()
else:
max_year = yt
# data contain more sectors since 2016
for year in range(yf, max_year + 1):
one_year_stocks = stocks.where(
cond=stocks['year'] == year).dropna()
sectors_month_stats = pd.DataFrame()
# print(one_year_stocks)
min_month = one_year_stocks.month.min()
max_month = one_year_stocks.month.max()
for month in range(int(min_month), int(max_month) + 1):
one_month_stocks = one_year_stocks.where(
cond=one_year_stocks['month'] == month).dropna()
one_month_stocks = Utils.add_first_last_perc_diff(
one_month_stocks)
sectors_data = one_month_stocks.groupby(
by=one_month_stocks['sector']).mean()
sectors_month_stats["sector"] = sectors_data.index
sectors_month_stats[month] = sectors_data.flpd.tolist()
sectors_month_stats.set_index('sector', inplace=True)
sec_year_stats[year] = sectors_month_stats.copy()
if show_chart:
self.show_sectors_stats(sec_year_stats, False, plt)
return sectors_month_stats
def top_stocks(self,
group=None,
ascending=False,
time_from=None,
time_to=None,
table=None,
from_top=0,
show_stocks_num=20,
is_industries=False,
separator=None):
st.empty()
table = TableName.DAY
if time_from.find("m") > -1:
table = TableName.MIN15
subject = "Loosers: " if ascending else "Gainers: "
if isinstance(group, str):
industry = str.split(group, separator)
group = industry[1] if len(industry) > 1 else industry[0]
group = [group]
if is_industries:
stocks = self.db.load_data(
table_name=table,
industries=group,
time_from=time_from,
time_to=time_to,
)
else:
stocks = self.db.load_data(
table_name=table,
sectors=group,
time_from=time_from,
time_to=time_to,
)
# self.stocks = FinI.add_change(self.stocks)
stocks = Utils.add_first_last_perc_diff(stocks)
self.print_stocks_list(stocks,
ascending,
from_top=from_top,
show_stocks_num=show_stocks_num)
