def stock_zh_index_daily_tx(self, stockcode):
if stockcode[0:1] == '6':
stockcode = 'sh' + stockcode
else:
stockcode = 'sz' + stockcode
stock_zh_index_daily_df = ak.stock_zh_index_daily_tx(symbol=stockcode)
return stock_zh_index_daily_df
def index_price_daily(code: str, source: str = None) -> dict:
if source == 'tx':
df = ak.stock_zh_index_daily_tx(symbol=code)
elif source == 'em':
df = ak.stock_zh_index_daily_em(symbol=code)
else:
df = ak.stock_zh_index_daily(symbol=code)
df['date'] = pd.to_datetime(df['date'])
df = df.rename({'date': '_id'}, axis=1)
print(df)
return json.loads(df.T.to_json()) # use default date_format='epoch' instead of 'iso'
def index_price_daily(code: str, source: str = None) -> pd.DataFrame:
if source == 'tx':
df = ak.stock_zh_index_daily_tx(symbol=code)
elif source == 'em':
df = ak.stock_zh_index_daily_em(symbol=code)
else:
df = ak.stock_zh_index_daily(symbol=code)
df['date'] = pd.to_datetime(df['date'])
df = df.rename({'date': '_id'}, axis=1)
print(df)
return df
def indexStocksCN(idx):
print(idx)
index_stock_cons_df = ak.index_stock_cons(index=idx[2:])
# print(index_stock_cons_df)
index_stock_cons_df.drop_duplicates(subset='品种代码',
keep='first',
inplace=True)
index_stock_cons_df.set_index('品种代码', inplace=True)
# 今日全场股票概况（价格、市值等）
df = ak.stock_zh_a_spot()
df.set_index('symbol', inplace=True)
# 按总市值排顺序，只保留指数成份股
df = df.loc[df.index.isin(
[dealSymbol(s) for s in index_stock_cons_df.index])].copy()
df.sort_values(by='mktcap', ascending=True, inplace=True)
df = df.iloc[:int(len(df) / 2) - 1]
# 计算因子值
days = 30
sortedDf = pd.DataFrame()
tqdmRange = tqdm(range(len(df)))
tqdmRange = tqdm([-3, -2, -1])
for i in tqdmRange:
symbol = dealSymbol(df.index[i])
tqdmRange.set_description("cauculating factor value for %s" % symbol)
factorVaule_1 = []
factorVaule_2 = []
stockK = ak.stock_zh_index_daily_tx(symbol=symbol)
stockK['symbol'] = symbol
for j in range(0, len(stockK)):
if j < days:
factorVaule_1.append(None)
factorVaule_2.append(None)
continue
factorVaule_1.append(factor_1(stockK.iloc[j - days:j]))
factorVaule_2.append(factor_1(stockK.iloc[j - days:j]))
stockK['factor_1'] = factorVaule_1
stockK['factor_2'] = factorVaule_2
stockK.to_csv('Quotation/' + symbol + '.csv')
stockK = stockK[['symbol', 'factor_1', 'factor_2']].copy()
stockK.reset_index(inplace=True)
sortedDf = sortedDf.append(stockK)
sortedDf.sort_values(by=['date', 'factor_1'], ascending=True, inplace=True)
sortedDf.dropna(subset=['factor_1'], inplace=True)
sortedDf.to_csv(idx + 'factor.csv')
sortedDf['date'] = pd.to_datetime(sortedDf['date']).date
return sortedDf
def get_kline(code, start_date: datetime.date):
"""
返回json数据，最新的日期在前面
"""
result = []
df = ak.stock_zh_a_daily(symbol=get_type(code) + code, adjust="hfq")
# 腾讯，主要是使用amount字段
dftx = ak.stock_zh_index_daily_tx(symbol=get_type(code) + code)
if df is None or df.empty:
return result
if dftx is None or dftx.empty:
return result
for row in df.itertuples():
w = {}
w['datetime'] = getattr(row, 'Index')
if w['datetime'] < start_date:
continue
w['open'] = getattr(row, 'open')
w['high'] = getattr(row, 'high')
w['low'] = getattr(row, 'low')
w['close'] = getattr(row, 'close')
w['volume'] = getattr(row, 'volume')
result.append(w)
return result
def get_60d_kline_data(s_code):
df = ak.stock_zh_index_daily_tx(symbol=f"sz{s_code}")
df.index = pd.to_datetime(df.index)
df_60d = df[-60:]
return df_60d
def research():
iK = dict()
idx = ['sh000300', 'sz399101', 'sz399102']
for iCode in idx:
iK[iCode] = ak.stock_zh_index_daily_tx(symbol=iCode)
num = min(len(iK['sh000300']), len(iK['sz399101']), len(iK['sz399102']))
for iCode in iK.keys():
iK[iCode] = iK[iCode].iloc[-num:]
iK[iCode][iCode] = [
iK[iCode]['close'][j] / iK[iCode]['close'][0]
for j in range(len(iK[iCode]['close']))
]
# 绘制走势
fig = iK[idx[0]][idx[0]].plot(figsize=(20, 5), color='purple')
fig.plot(iK[idx[1]][idx[1]], color='green', alpha=1)
fig.plot(iK[idx[2]][idx[2]], color='brown', alpha=1)
days = 20
i = -days
signal = 0
nSeri = []
while i < len(iK[idx[0]]) - days - 1:
i += days
if i < days:
nSeri.extend([1.0 for r in range(days + 1)])
continue
# 收益曲线
compare = [
iK[idx[0]][idx[0]][i] / iK[idx[0]][idx[0]][i - days],
iK[idx[1]][idx[1]][i] / iK[idx[1]][idx[1]][i - days],
iK[idx[2]][idx[2]][i] / iK[idx[2]][idx[2]][i - days]
]
m = compare.index(max(compare))
nSeri.extend([
nSeri[-1] * iK[idx[m]][idx[m]][r] / iK[idx[m]][idx[m]][i - days]
for r in range(i - days, i)
])
fig.plot(pd.Series(data=nSeri, index=iK[idx[m]].index),
color='black',
alpha=1)
fig.legend()
timespans = {'x1': [], 'x2': [], 'o1': [], 'o2': []}
colors = {'x1': 'green', 'x2': 'blue', 'o1': 'red', 'o2': 'orange'}
cs = iK['sh000300']['close']
i = -1
days = 44
x2 = timespans['x2']
while i < len(cs) - 1 - days:
i += 1
c = cs[i:i + days]
if c[-1] < c[0] and max(c) == c[0]:
i += days - 1
x2.append(c.index)
days = 32
spans = []
testnum = 4
for n in tqdm(range(testnum)):
nSpan = []
i = -1
while i < len(cs) - 1 - days:
i += 1
c = cs[i:i + days]
if int(days / testnum * n) <= c.values.argmax() < int(
days / testnum * (n + 1)) and c[-1] < c[0]:
nSpan.append(c.index)
i += days - 1
spans.append(nSpan)
minCount = 999
minSpaniK = 0
for si in range(testnum):
match = 0
for sp in spans[si]:
for ts in timespans['x2']:
if ts[0] in sp:
match += 1
if len(spans[si]) < minCount and len(spans[si]) != 0 and match > 0:
minCount = len(spans[si])
minSpaniK = si
timespans['x1'] = spans[minSpaniK]
print(len(spans[minSpaniK]), minSpaniK)
for k, v in timespans.items():
if len(v) == 0:
continue
for span in v:
if len(span) == 0:
continue
fig.axvspan(span[0],
span[-1],
facecolor=colors[k],
edgecolor='none',
alpha=.2)
plt.show()
def stock_zh_index_daily_tx(self):
stock_zh_index_daily_tx_df = ak.stock_zh_index_daily_tx(
symbol="sh000919")
return stock_zh_index_daily_tx_df