akshare 1.14.49__py3-none-any.whl → 1.17.99__py3-none-any.whl

akshare/cal/rv.py

@@ -0,0 +1,170 @@
+"""
+Yang-Zhang-s-Realized-Volatility-Automated-Estimation-in-Python
+https://github.com/hugogobato/Yang-Zhang-s-Realized-Volatility-Automated-Estimation-in-Python
+论文地址：https://www.jstor.org/stable/10.1086/209650
+"""
+
+import warnings
+
+import numpy as np
+import pandas as pd
+
+
+def rv_from_stock_zh_a_hist_min_em(
+    symbol="000001",
+    start_date="2021-10-20 09:30:00",
+    end_date="2024-11-01 15:00:00",
+    period="1",
+    adjust="hfq",
+) -> pd.DataFrame:
+    """
+    从东方财富网获取股票的分钟级历史行情数据,并进行数据清洗和格式化为计算 yz 已实现波动率所需的数据格式
+    https://quote.eastmoney.com/concept/sh603777.html?from=classic
+    :param symbol: 股票代码,如"000001"
+    :type symbol: str
+    :param start_date: 开始日期时间,格式"YYYY-MM-DD HH:MM:SS"
+    :type start_date: str
+    :param end_date: 结束日期时间,格式"YYYY-MM-DD HH:MM:SS"
+    :type end_date: str
+    :param period: 时间周期,可选{'1','5','15','30','60'}分钟
+    :type period: str
+    :param adjust: 复权方式,可选{'','qfq'(前复权),'hfq'(后复权)}
+    :type adjust: str
+    :return: 整理后的分钟行情数据,包含Date(索引),Open,High,Low,Close列
+    :rtype: pandas.DataFrame
+    """
+    from akshare.stock_feature.stock_hist_em import stock_zh_a_hist_min_em
+
+    temp_df = stock_zh_a_hist_min_em(
+        symbol=symbol,
+        start_date=start_date,
+        end_date=end_date,
+        period=period,
+        adjust=adjust,
+    )
+    temp_df.rename(
+        columns={
+            "时间": "Date",
+            "开盘": "Open",
+            "最高": "High",
+            "最低": "Low",
+            "收盘": "Close",
+        },
+        inplace=True,
+    )
+    temp_df = temp_df[temp_df["Open"] != 0]
+    temp_df["Date"] = pd.to_datetime(temp_df["Date"])
+    temp_df.set_index(keys="Date", inplace=True)
+    return temp_df
+
+
+def rv_from_futures_zh_minute_sina(
+    symbol: str = "IF2008", period: str = "5"
+) -> pd.DataFrame:
+    """
+    从新浪财经获取期货的分钟级历史行情数据,并进行数据清洗和格式化
+    https://vip.stock.finance.sina.com.cn/quotes_service/view/qihuohangqing.html#titlePos_3
+    :param symbol: 期货合约代码,如"IF2008"代表沪深300期货2020年8月合约
+    :type symbol: str
+    :param period: 时间周期,可选{'1','5','15','30','60'}分钟
+    :type period: str
+    :return: 整理后的分钟行情数据,包含Date(索引),Open,High,Low,Close列
+    :rtype: pandas.DataFrame
+    """
+    from akshare.futures.futures_zh_sina import futures_zh_minute_sina
+
+    temp_df = futures_zh_minute_sina(symbol=symbol, period=period)
+    temp_df.rename(
+        columns={
+            "datetime": "Date",
+            "open": "Open",
+            "high": "High",
+            "low": "Low",
+            "close": "Close",
+        },
+        inplace=True,
+    )
+    temp_df["Date"] = pd.to_datetime(temp_df["Date"])
+    temp_df.set_index(keys="Date", inplace=True)
+    return temp_df
+
+
+def volatility_yz_rv(data: pd.DataFrame) -> pd.DataFrame:
+    """
+    波动率-已实现波动率-Yang-Zhang 已实现波动率(Yang-Zhang Realized Volatility)
+    https://github.com/hugogobato/Yang-Zhang-s-Realized-Volatility-Automated-Estimation-in-Python
+    论文地址：https://www.jstor.org/stable/10.1086/209650
+    基于以下公式计算:
+    RV^2 = Vo + k*Vc + (1-k)*Vrs
+    其中:
+    - Vo: 隔夜波动率, Vo = 1/(n-1)*sum(Oi-Obar)^2
+        Oi为标准化开盘价, Obar为标准化开盘价均值
+    - Vc: 收盘波动率, Vc = 1/(n-1)*sum(ci-Cbar)^2
+        ci为标准化收盘价, Cbar为标准化收盘价均值
+    - k: 权重系数, k = 0.34/(1.34+(n+1)/(n-1))
+        n为样本数量
+    - Vrs: Rogers-Satchell波动率代理, Vrs = ui(ui-ci)+di(di-ci)
+        ui = ln(Hi/Oi), ci = ln(Ci/Oi), di = ln(Li/Oi), oi = ln(Oi/Ci-1)
+        Hi/Li/Ci/Oi分别为最高价/最低价/收盘价/开盘价
+
+    :param data: 包含 OHLC(开高低收) 价格的 pandas.DataFrame
+    :type data: pandas.DataFrame
+    :return: 包含 Yang-Zhang 实现波动率的 pandas.DataFrame
+    :rtype: pandas.DataFrame
+
+    要求输入数据包含以下列:
+    - Open: 开盘价
+    - High: 最高价
+    - Low: 最低价
+    - Close: 收盘价
+    # yang_zhang_rv formula is give as:
+    # RV^2 = Vo + k*Vc + (1-k)*Vrs
+    # where Vo = 1/(n-1)*sum(Oi-Obar)^2
+    # with oi = normalized opening price at time t and Obar = mean of normalized opening prices
+    # Vc = = 1/(n-1)*sum(ci-Cbar)^2
+    # with ci = normalized close price at time t and Cbar = mean of normalized close prices
+    # k = 0.34/(1.34+(n+1)/(n-1))
+    # with n = total number of days or time periods considered
+    # Vrs (Rogers & Satchell RV proxy) = ui(ui-ci)+di(di-ci)
+    # with ui = ln(Hi/Oi), ci = ln(Ci/Oi), di=(Li/Oi), oi = ln(Oi/Ci-1)
+    # where Hi = high price at time t and Li = low price at time t
+    """ ""
+    warnings.filterwarnings("ignore")
+
+    data["ui"] = np.log(np.divide(data["High"][1:], data["Open"][1:]))
+    data["ci"] = np.log(np.divide(data["Close"][1:], data["Open"][1:]))
+    data["di"] = np.log(np.divide(data["Low"][1:], data["Open"][1:]))
+    data["oi"] = np.log(np.divide(data["Open"][1:], data["Close"][: len(data) - 1]))
+    data = data[1:]
+    data["RS"] = data["ui"] * (data["ui"] - data["ci"]) + data["di"] * (
+        data["di"] - data["ci"]
+    )
+    rs_var = data["RS"].groupby(pd.Grouper(freq="1D")).mean().dropna()
+    vc_and_vo = data[["oi", "ci"]].groupby(pd.Grouper(freq="1D")).var().dropna()
+    n = int(len(data) / len(rs_var))
+    k = 0.34 / (1.34 + (n + 1) / (n - 1))
+    yang_zhang_rv = np.sqrt((1 - k) * rs_var + vc_and_vo["oi"] + vc_and_vo["ci"] * k)
+    yang_zhang_rv_df = pd.DataFrame(yang_zhang_rv)
+    yang_zhang_rv_df.rename(columns={0: "yz_rv"}, inplace=True)
+    yang_zhang_rv_df.reset_index(inplace=True)
+    yang_zhang_rv_df.columns = ["date", "rv"]
+    yang_zhang_rv_df["date"] = pd.to_datetime(
+        yang_zhang_rv_df["date"], errors="coerce"
+    ).dt.date
+    return yang_zhang_rv_df
+
+
+if __name__ == "__main__":
+    futures_df = rv_from_futures_zh_minute_sina(symbol="IF2008", period="1")
+    volatility_yz_rv_df = volatility_yz_rv(data=futures_df)
+    print(volatility_yz_rv_df)
+
+    stock_df = rv_from_stock_zh_a_hist_min_em(
+        symbol="000001",
+        start_date="2021-10-20 09:30:00",
+        end_date="2024-11-01 15:00:00",
+        period="5",
+        adjust="",
+    )
+    volatility_yz_rv_df = volatility_yz_rv(data=stock_df)
+    print(volatility_yz_rv_df)

akshare/cost/cost_living.py

@@ -1,10 +1,11 @@
 #!/usr/bin/env python
 # -*- coding:utf-8 -*-
 """
-Date: 2024/1/20 23:30
+Date: 2024/8/28 15:20
 Desc: 世界各大城市生活成本数据
 https://expatistan.com/cost-of-living/index
 """
+
 from io import StringIO
 
 import pandas as pd
@@ -20,14 +21,14 @@ def _get_region() -> dict:
     """
     url = "https://www.expatistan.com/cost-of-living/index"
     r = requests.get(url)
-    soup = BeautifulSoup(r.text, "lxml")
+    soup = BeautifulSoup(r.text, features="lxml")
     half_url_list = [
         item["href"]
-        for item in soup.find("ul", attrs={"class": "regions"}).find_all("a")
+        for item in soup.find(name="ul", attrs={"class": "regions"}).find_all("a")
     ]
     name_list = [
         item["href"].split("/")[-1]
-        for item in soup.find("ul", attrs={"class": "regions"}).find_all("a")
+        for item in soup.find(name="ul", attrs={"class": "regions"}).find_all("a")
     ]
     name_url_dict = dict(zip(name_list, half_url_list))
     name_url_dict["world"] = "/cost-of-living/index"
@@ -38,7 +39,8 @@ def cost_living(symbol: str = "world") -> pd.DataFrame:
     """
     国家或地区生活成本数据
     https://expatistan.com/cost-of-living/index
-    :param symbol: choice of {"europe", "north-america", "latin-america", "asia", "middle-east", "africa", "oceania", "world"}
+    :param symbol: choice of {"europe", "north-america", "latin-america",
+    "asia", "middle-east", "africa", "oceania", "world"}
     :type symbol: str
     :return: 国家或地区生活成本数据
     :rtype: pandas.DataFrame

akshare/crypto/__init__.py

@@ -2,5 +2,5 @@
 # -*- coding:utf-8 -*-
 """
 Date: 2020/10/23 13:51
-Desc: 
+Desc:
 """

akshare/crypto/crypto_bitcoin_cme.py

@@ -5,6 +5,7 @@ Date: 2023/9/5 15:41
 Desc: 芝加哥商业交易所-比特币成交量报告
 https://datacenter.jin10.com/reportType/dc_cme_btc_report
 """
+
 import pandas as pd
 import requests
 
@@ -13,6 +14,8 @@ def crypto_bitcoin_cme(date: str = "20230830") -> pd.DataFrame:
     """
     芝加哥商业交易所-比特币成交量报告
     https://datacenter.jin10.com/reportType/dc_cme_btc_report
+    :param date: Specific date, e.g., "20230830"
+    :type date: str
     :return: 比特币成交量报告
     :rtype: pandas.DataFrame
     """
@@ -21,7 +24,6 @@ def crypto_bitcoin_cme(date: str = "20230830") -> pd.DataFrame:
         "category": "cme",
         "date": "-".join([date[:4], date[4:6], date[6:]]),
         "attr_id": "4",
-        "_": "1624354777843",
     }
     headers = {
         "accept": "*/*",
@@ -47,12 +49,12 @@ def crypto_bitcoin_cme(date: str = "20230830") -> pd.DataFrame:
         [item for item in data_json["data"]["values"]],
         columns=[item["name"] for item in data_json["data"]["keys"]],
     )
-    temp_df['电子交易合约'] = pd.to_numeric(temp_df['电子交易合约'], errors="coerce")
-    temp_df['场内成交合约'] = pd.to_numeric(temp_df['场内成交合约'], errors="coerce")
-    temp_df['场外成交合约'] = pd.to_numeric(temp_df['场外成交合约'], errors="coerce")
-    temp_df['成交量'] = pd.to_numeric(temp_df['成交量'], errors="coerce")
-    temp_df['未平仓合约'] = pd.to_numeric(temp_df['未平仓合约'], errors="coerce")
-    temp_df['持仓变化'] = pd.to_numeric(temp_df['持仓变化'], errors="coerce")
+    temp_df["电子交易合约"] = pd.to_numeric(temp_df["电子交易合约"], errors="coerce")
+    temp_df["场内成交合约"] = pd.to_numeric(temp_df["场内成交合约"], errors="coerce")
+    temp_df["场外成交合约"] = pd.to_numeric(temp_df["场外成交合约"], errors="coerce")
+    temp_df["成交量"] = pd.to_numeric(temp_df["成交量"], errors="coerce")
+    temp_df["未平仓合约"] = pd.to_numeric(temp_df["未平仓合约"], errors="coerce")
+    temp_df["持仓变化"] = pd.to_numeric(temp_df["持仓变化"], errors="coerce")
     return temp_df
 
 

akshare/crypto/crypto_hold.py

@@ -5,6 +5,7 @@ Date: 2023/8/31 23:00
 Desc: 金十数据-比特币持仓报告
 https://datacenter.jin10.com/dc_report?name=bitcoint
 """
+
 import pandas as pd
 import requests
 
@@ -22,7 +23,6 @@ def crypto_bitcoin_hold_report():
         "X-Version": "1.0.0",
     }
     r = requests.get(url, headers=headers)
-
     data_json = r.json()
     temp_df = pd.DataFrame(data_json["data"]["values"])
     temp_df.columns = [
@@ -62,7 +62,9 @@ def crypto_bitcoin_hold_report():
         ]
     ]
     temp_df["市值"] = pd.to_numeric(temp_df["市值"], errors="coerce")
-    temp_df["比特币占市值比重"] = pd.to_numeric(temp_df["比特币占市值比重"], errors="coerce")
+    temp_df["比特币占市值比重"] = pd.to_numeric(
+        temp_df["比特币占市值比重"], errors="coerce"
+    )
     temp_df["持仓成本"] = pd.to_numeric(temp_df["持仓成本"], errors="coerce")
     temp_df["持仓占比"] = pd.to_numeric(temp_df["持仓占比"], errors="coerce")
     temp_df["持仓量"] = pd.to_numeric(temp_df["持仓量"], errors="coerce")

akshare/currency/currency.py

@@ -6,6 +6,7 @@ Desc: currencybeacon 提供的外汇数据
 该网站需要先注册后获取 API 使用
 https://currencyscoop.com/
 """
+
 import pandas as pd
 import requests
 

akshare/currency/currency_china_bank_sina.py

@@ -1,10 +1,11 @@
 #!/usr/bin/env python
 # -*- coding:utf-8 -*-
 """
-Date: 2023/11/10 15:20
+Date: 2025/12/8 17:20
 Desc: 新浪财经-中行人民币牌价历史数据查询
 https://biz.finance.sina.com.cn/forex/forex.php?startdate=2012-01-01&enddate=2021-06-14&money_code=EUR&type=0
 """
+
 from functools import lru_cache
 from io import StringIO
 
@@ -16,7 +17,7 @@ from tqdm import tqdm
 
 @lru_cache()
 def _currency_boc_sina_map(
-        start_date: str = "20210614", end_date: str = "20230810"
+    start_date: str = "20210614", end_date: str = "20230810"
 ) -> dict:
     """
     外汇 symbol 和代码映射
@@ -54,7 +55,7 @@ def _currency_boc_sina_map(
 
 
 def currency_boc_sina(
-        symbol: str = "美元", start_date: str = "20230304", end_date: str = "20231110"
+    symbol: str = "美元", start_date: str = "20230304", end_date: str = "20231110"
 ) -> pd.DataFrame:
     """
     新浪财经-中行人民币牌价历史数据查询
@@ -79,7 +80,7 @@ def currency_boc_sina(
         "call_type": "ajax",
     }
     r = requests.get(url, params=params)
-    soup = BeautifulSoup(r.text, "lxml")
+    soup = BeautifulSoup(r.text, features="lxml")
     soup.find(attrs={"id": "money_code"})
     page_element_list = soup.find_all("a", attrs={"class": "page"})
     page_num = int(page_element_list[-2].text) if len(page_element_list) != 0 else 1
@@ -88,19 +89,23 @@ def currency_boc_sina(
         params.update({"page": page})
         r = requests.get(url, params=params)
         temp_df = pd.read_html(StringIO(r.text), header=0)[0]
-        big_df = pd.concat([big_df, temp_df], ignore_index=True)
+        big_df = pd.concat(objs=[big_df, temp_df], ignore_index=True)
     big_df.columns = [
         "日期",
         "中行汇买价",
         "中行钞买价",
         "中行钞卖价/汇卖价",
         "央行中间价",
+        "中行折算价",
     ]
     big_df["日期"] = pd.to_datetime(big_df["日期"], errors="coerce").dt.date
     big_df["中行汇买价"] = pd.to_numeric(big_df["中行汇买价"], errors="coerce")
     big_df["中行钞买价"] = pd.to_numeric(big_df["中行钞买价"], errors="coerce")
-    big_df["中行钞卖价/汇卖价"] = pd.to_numeric(big_df["中行钞卖价/汇卖价"], errors="coerce")
+    big_df["中行钞卖价/汇卖价"] = pd.to_numeric(
+        big_df["中行钞卖价/汇卖价"], errors="coerce"
+    )
     big_df["央行中间价"] = pd.to_numeric(big_df["央行中间价"], errors="coerce")
+    big_df["中行折算价"] = pd.to_numeric(big_df["中行折算价"], errors="coerce")
     big_df.sort_values(by=["日期"], inplace=True, ignore_index=True)
     return big_df
 

akshare/data/__init__.py

@@ -2,5 +2,5 @@
 # !/usr/bin/env python
 """
 Date: 2022/5/9 18:08
-Desc: 
+Desc:
 """

akshare/data/cninfo.js

@@ -6013,4 +6013,4 @@ function new_my_decode(data) {
         }
     );
     return decrypted.toString(CryptoJS.enc.Utf8);
-}
+}

akshare/datasets.py

@@ -1,24 +1,19 @@
 # -*- coding:utf-8 -*-
 # !/usr/bin/env python
 """
-Date: 2023/9/24 16:08
+Date: 2024/12/30 15:30
 Desc: 导入文件工具，可以正确处理路径问题
 """
+
 import pathlib
 from importlib import resources
 
 
 def get_ths_js(file: str = "ths.js") -> pathlib.Path:
-    """Get path to data "ths.js" text file.
-
-    Returns
-    -------
-    pathlib.PosixPath
-        Path to file.
-
-    References
-    ----------
-    .. [1] E.A.Abbott, ”Flatland”, Seeley & Co., 1884.
+    """
+    get path to data "ths.js" text file.
+    :return: 文件路径
+    :rtype: pathlib.Path
     """
     with resources.path("akshare.data", file) as f:
         data_file_path = f
@@ -26,16 +21,10 @@ def get_ths_js(file: str = "ths.js") -> pathlib.Path:
 
 
 def get_crypto_info_csv(file: str = "crypto_info.zip") -> pathlib.Path:
-    """Get path to data "ths.js" text file.
-
-    Returns
-    -------
-    pathlib.PosixPath
-        Path to file.
-
-    References
-    ----------
-    .. [1] E.A.Abbott, ”Flatland”, Seeley & Co., 1884.
+    """
+    get path to data "ths.js" text file.
+    :return: 文件路径
+    :rtype: pathlib.Path
     """
     with resources.path("akshare.data", file) as f:
         data_file_path = f

akshare/economic/cons.py

@@ -4,6 +4,7 @@
 Date: 2019/10/21 21:11
 Desc: 宏观经济配置文件
 """
+
 # urls-china
 JS_CHINA_CPI_YEARLY_URL = (
     "https://cdn.jin10.com/dc/reports/dc_chinese_cpi_yoy_all.js?v={}&_={}"
@@ -24,7 +25,9 @@ JS_CHINA_GDP_YEARLY_URL = (
     "https://cdn.jin10.com/dc/reports/dc_chinese_gdp_yoy_all.js?v={}&_={}"
 )
 JS_CHINA_CX_PMI_YEARLY_URL = "https://cdn.jin10.com/dc/reports/dc_chinese_caixin_manufacturing_pmi_all.js?v={}&_={}"
-JS_CHINA_CX_SERVICE_PMI_YEARLY_URL = "https://cdn.jin10.com/dc/reports/dc_chinese_caixin_services_pmi_all.js?v={}&_={}"
+JS_CHINA_CX_SERVICE_PMI_YEARLY_URL = (
+    "https://cdn.jin10.com/dc/reports/dc_chinese_caixin_services_pmi_all.js?v={}&_={}"
+)
 JS_CHINA_FX_RESERVES_YEARLY_URL = (
     "https://cdn.jin10.com/dc/reports/dc_chinese_fx_reserves_all.js?v={}&_={}"
 )
@@ -35,8 +38,12 @@ JS_CHINA_NON_MAN_PMI_MONTHLY_URL = (
     "https://cdn.jin10.com/dc/reports/dc_chinese_non_manufacturing_pmi_all.js?v={}&_={}"
 )
 JS_CHINA_RMB_DAILY_URL = "https://cdn.jin10.com/dc/reports/dc_rmb_data_all.js?v={}&_={}"
-JS_CHINA_MARKET_MARGIN_SZ_URL = "https://cdn.jin10.com/dc/reports/dc_market_margin_sz_all.js?v={}&_={}"
-JS_CHINA_MARKET_MARGIN_SH_URL = "https://cdn.jin10.com/dc/reports/dc_market_margin_sse_all.js?v={}&_={}"
+JS_CHINA_MARKET_MARGIN_SZ_URL = (
+    "https://cdn.jin10.com/dc/reports/dc_market_margin_sz_all.js?v={}&_={}"
+)
+JS_CHINA_MARKET_MARGIN_SH_URL = (
+    "https://cdn.jin10.com/dc/reports/dc_market_margin_sse_all.js?v={}&_={}"
+)
 JS_CHINA_REPORT_URL = "https://cdn.jin10.com/dc/reports/dc_sge_report_all.js?v={}&_={}"
 
 # urls-usa