{"id":3741,"date":"2022-03-18T12:37:33","date_gmt":"2022-03-18T04:37:33","guid":{"rendered":"https:\/\/egonlin.com\/?p=3741"},"modified":"2022-03-18T12:37:33","modified_gmt":"2022-03-18T04:37:33","slug":"%e7%ac%ac%e4%b8%80%e8%8a%82%ef%bc%9a%e7%bb%9f%e8%ae%a1%e5%88%86%e6%9e%90","status":"publish","type":"post","link":"https:\/\/egonlin.com\/?p=3741","title":{"rendered":"\u7b2c\u4e00\u8282\uff1a\u7edf\u8ba1\u5206\u6790"},"content":{"rendered":"

\u7edf\u8ba1\u5206\u6790<\/h2>\n

\"\"<\/div><\/p>\n

1\u3001\u8ba1\u7b97\u56de\u62a5\u7387<\/h4>\n

\u200b \u901a\u5e38\u5728\u91d1\u878d\u884c\u4e1a\u4f1a\u7528\u65e5\u56de\u62a5\u7387\u6765\u8ba1\u7b97\u8d1d\u5854\u503c\uff0c\u4ee5\u8861\u91cf\u4e00\u5bb6\u516c\u53f8\u7684\u5e02\u573a\u98ce\u9669\u3002\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u4ef7\u683c\u6570\u636e\u6765\u8ba1\u7b97\u56de\u62a5\u7387\uff08\u4e5f\u79f0\u4f5c\u6536\u76ca\u7387\uff09\uff0c\u6709\u65f6\u4f1a\u9700\u8981\u5c06\u65e5\u6536\u76ca\u6bcf\u5468\u6216\u8005\u6bcf\u6708\u6536\u76ca\u7387\uff0c\u6240\u4ee5\u63a5\u4e0b\u6765\u4e3b\u8981\u4ecb\u7ecd\u7684\u5c31\u662f\u6536\u76ca\u7387\u7684\u8ba1\u7b97\u3002\u8ba1\u7b97\u56de\u62a5\u7387\u6700\u91cd\u8981\u7684\u5c31\u662f\u8981\u660e\u767d\u8fd9\u4e9b\u4ef7\u683c\u7684\u6392\u5e8f\u3002<\/p>\n

import pandas as pd\nimport tushare as ts\n\ndata = ts.get_k_data('000001')  # \u83b7\u53d6\u5e73\u5b89\u7684k\u7ebf\u6570\u636e\nclose = np.array(data.close)  # \u8f6c\u6362\u4e3andarray\u6570\u7ec4\uff0c\u4ee5\u4fbf\u505a\u6279\u91cf\u8ba1\u7b97\nret  = (close[1:]-close[:-1])\/close[:-1]\nret[0:2]\n\u8fd0\u884c\u7ed3\u679c\uff1a\narray([0.00775724, 0.00441476])\n\n# \u9a8c\u8bc1\u56de\u62a5\u7387\uff1a\n> data.date[0:3]\n>:\n0    2016-08-05\n1    2016-08-08\n2    2016-08-09\nName: date, dtype: object\n\n> data.close[0:3]\n>:\n0    8.766\n1    8.834\n2    8.873\nName: close, dtype: float64\n\n> (8.834-8.766)\/8.766\n> 0.007757243896874243<\/code><\/pre>\n
1.1\u3001\u4ece\u65e5\u56de\u62a5\u7387\u8ba1\u7b97\u6708\u56de\u62a5\u7387<\/h5>\n
$P_0$\u4e3a\u4e0a\u4e2a\u6708\u6700\u540e\u4e00\u4e2a\u4ea4\u6613\u65e5\u7684\u6536\u76d8\u4ef7\uff0c$P_1$\u662f\u8fd9\u4e2a\u6708\u7b2c\u4e00\u4e2a\u4ea4\u6613\u65e5\u7684\u6536\u76d8\u4ef7\u3002<\/p>\n
\u6708\u56de\u62a5\u7387\uff1a
\n$$
\nR{monthly} = \\frac {P<\/em>{20} – P_0} {P0}
\n$$
\n\u6708\u5bf9\u6570\u56de\u62a5\u7387\uff1a
\n$$
\nlogreturn<\/em>{monthly} = log({\\frac {P_{20}}{P_0}})
\n$$
\n\u65e5\u5bf9\u6570\u56de\u62a5\u7387\uff1a
\n$$
\nlogreturn^{daily}_i = log({\\frac{Pi}{P<\/em>{i-1}}})
\n$$
\n\u767e\u5206\u6bd4\u56de\u62a5\u7387\u548c\u5bf9\u6570\u56de\u62a5\u7387\u4e4b\u95f4\u7684\u5173\u7cfb\uff1a
\n$$
\nR{monthly} = exp(logreturn) – 1
\n$$
\n\u5bf9\u6570\u56de\u62a5\u7387\u7684\u603b\u548c\uff1a
\n$$
\nlogreturn<\/em>{monthly} = log({\\frac{P_{20}}{P0}}) = \\displaystyle \\sum^{20}<\/em>{i=1}logreturn^{daily}_i
\n$$<\/p>\n
# \u5bf9\u6570\u56de\u62a5\u7387\nlogret = np.log(close[1:]\/close[:-1])\n\n# \u5904\u7406\u65f6\u95f4\u683c\u5f0f\ndd = []\nfor dt in data.date:\n    # \u5b57\u7b26\u4e32\u683c\u5f0f\u7684\u65f6\u95f4\u5bf9\u8c61\u53ea\u6709\u5148\u8f6c\u6362\u4e3adatetime\u683c\u5f0f\u624d\u80fd\u8fdb\u884c\u8f6c\u6362\u5230date\u683c\u5f0f\n    d0 = dt.strptime(i,'%Y-%m-%d')  # \u8f6c\u6362\u4e3adatetime.datetime\u683c\u5f0f\n    d1 = d0.date()  # \u8f6c\u6362\u4e3adatetime.date\u683c\u5f0f\n    dd.append(d1)\n# \u5c06\u65b0\u5217\u8868\u8f6c\u6362\u4e3andarray\u6570\u7ec4\nda = np.array(dd)\n\n# \u6708\u56de\u62a5\u7387\nym = []\nfor i in range(0,np.size(logret)):\n    ym.append(''.join([da[i].strftime("%Y"),da[i].strftime("%m")]))\n# \u901a\u8fc7pandas\u8fdb\u884c\u6279\u91cf\u8ba1\u7b97\ny = pd.DataFrame(logret,ym,columns=['ret_monthly'])\nret_monthly = y.groupby(y.index).sum()\nret_monthly\n\u8fd0\u884c\u7ed3\u679c:\n        ret_monthly\n201608  0.044402\n201609  -0.035544\n201610  0.002146\n201611  0.049096\n201612  -0.046842\n...\n201811  -0.022410\n201812  -0.141794\n201901  0.197798\n201902  0.130401\n201903  -0.053110<\/code><\/pre>\n
\u4ee5\u4e0a\u7a0b\u5e8f\u4f9d\u65e7\u901a\u8fc7\u83b7\u53d6\u7684\u5e73\u5b89\u7684K\u7ebf\u6570\u636e\u6765\u8fdb\u884c\u8ba1\u7b97<\/p>\n
1.2\u3001\u4ece\u65e5\u56de\u62a5\u7387\u8ba1\u7b97\u5e74\u56de\u62a5\u7387<\/h5>\n
\u65b9\u6cd5\u7c7b\u4f3c<\/p>\n
# \u5e74\u56de\u62a5\u7387\ndate = []\nfor i in range(0,np.size(logret)):\n    date.append(''.join([da[i].strftime("%Y")]))\ny = pd.DataFrame(logret,date,columns=['ret_annual'])\nret_annual = np.exp(y.groupby(y.index).sum())-1\n\u8fd0\u884c\u7ed3\u679c:\n        ret_annual\n2016    0.013347\n2017    0.517730\n2018    -0.318350\n2019    0.316649<\/code><\/pre>\n
2\u3001T-\u68c0\u9a8c<\/h4>\n
\"\"<\/div><\/p>\n
\u91d1\u878d\u9886\u57df\u5e38\u7528t-\u7edf\u8ba1\u91cf\u6765\u68c0\u9a8c\u67d0\u4e2a\u96f6\u5047\u8bbe\u3002\u5f53\u96f6\u5047\u8bbe\u6210\u7acb\u65f6\uff0ct-\u7edf\u8ba1\u91cf\u670d\u4ecet\u5206\u5e03\u3002<\/p>\n
\u4ee5\u4e0b\u7684\u4ee3\u7801\u751f\u62101000\u4e2a\u670d\u4ece\u6807\u51c6\u6b63\u6001\u5206\u5e03\u7684\u968f\u673a\u6570\uff0c\u7136\u540e\u8fdb\u884c\u4e24\u4e2a\u68c0\u9a8c\u3002\u7b2c\u4e00\u4e2a\u68c0\u9a8c\u7684\u96f6\u5047\u8bbe\u662f\u5747\u503c\u4e3a0.5\uff0c\u7b2c2\u4e2a\u68c0\u9a8c\u7684\u96f6\u5047\u8bbe\u662f\u5747\u503c\u4e3a0.<\/p>\n
from scipy import stats\nimport scipy as sp\nimport tushare as ts\nimport numpy as np\n\nsp.random.seed(1235)\nx = stats.norm.rvs(size=1000)\nplt.figure(figsize=(7,4))\n\n# \u5206\u522b\u4e3a\u4e24\u6761\u6570\u636e\u6dfb\u52a0\u56fe\u4f8b\nplt.plot(x,lw=1.0,label='1st')  \nplt.grid(True)  # \u7f51\u683c\u8bbe\u7f6e\nplt.legend(loc=0)  # \u56fe\u4f8b\u6807\u7b7e\u4f4d\u7f6e\u8bbe\u7f6e\nplt.axis('tight')\nplt.xlabel('index')\nplt.ylabel('value')<\/code><\/pre>\n
\u56fe\u4f8b:<\/p>\n
\"\"<\/div><\/p>\n
print(stats.ttest_1samp(x,0.5))\nprint(stats.ttest_1samp(x,0))\n\n\u8fd0\u884c\u7ed3\u679c:\nTtest_1sampResult(statistic=-16.755214405175575, pvalue=1.0140316700193316e-55)\nTtest_1sampResult(statistic=-1.1953686865174769, pvalue=0.23222669336297136)<\/code><\/pre>\n
\u6d4b\u8bd5\u4e2d\u56fd\u5e73\u5b89\u7684\u5e73\u5747\u65e5\u56de\u62a5\u7387\u662f\u5426\u4e3a0<\/p>\n
data = ts.get_k_data('000001')\npa_data_close = np.array(data.close)\nret = pa_data_close[1:]\/pa_data_close[:-1] - 1\n\nprint(round(sp.mean(ret),5),stats.ttest_1samp(ret,0))\n\n\u8fd0\u884c\u7ed3\u679c:\n0.00065 Ttest_1sampResult(statistic=0.9167715496440845, pvalue=0.35960876859507807)<\/code><\/pre>\n
\u7ed3\u679c\u663e\u793a\u4e2d\u56fd\u5e73\u5b89\u7684\u5e73\u5747\u65e5\u56de\u62a5\u7387\u4e3a0.065%\uff0ct\u503c\u662f0.92\u800cp\u503c\u662f0.36\u3002\u6240\u4ee5\uff0c\u7edf\u8ba1\u68c0\u9a8c\u7684\u7ed3\u679c\u4e0d\u652f\u6301\u5e73\u5747\u65e5\u56de\u62a5\u7387\u7b49\u4e8e0.<\/p>\n
3\u3001\u57fa\u4e8e52\u5468\u6700\u9ad8\u4ef7\u548c\u6700\u4f4e\u4ef7\u7684\u4ea4\u6613\u7b56\u7565<\/h4>\n
\"\"<\/div><\/p>\n
\u200b   \u5f53\u67d0\u53ea\u80a1\u7968\u4eca\u5929\u7684\u4ef7\u683c\u63a5\u8fd1\u8fc7\u53bb52\u5468\u7684\u6700\u9ad8\u4ef7\u683c\uff0c\u5356\u51fa\u8be5\u80a1\u7968\uff1b\u5f53\u5176\u4eca\u5929\u5f97\u4ef7\u683c\u63a5\u8fd1\u8fc7\u53bb52\u5468\u8fbe\u5230\u7684\u6700\u4f4e\u4ef7\u683c\uff0c\u4e70\u5165\u8be5\u80a1\u7968\u3002<\/p>\n
import tushare as ts\nfrom datetime import datetime\nimport numpy as np\n\nwk_data = ts.get_k_data("000002",start='2018-01-01',end='2018-12-30')  # \u83b7\u53d6\u4e07\u79d1\u5730\u4ea72018\u4e00\u5e74\u7684K\u7ebf\u6570\u636e\nwk_data = np.array(wk_data)  # \u8f6c\u6362\u4e3andarray\u6570\u7ec4\nwk_data_1 = wk_data[-1]\n\ny = np.array(wk_data.tolist())[:,-3]  # \u6700\u4f4e\u4ef7\naa = y.astype(np.float)  \nhigh = max(aa)  #  \u7b97\u51fa\u4e00\u5e74\u4e2d\u6700\u9ad8\u7684\nlow = min(aa)  #  \u7b97\u51fa\u4e00\u5e74\u4e2d\u6700\u4f4e\u7684\n\nprint("""\ntoday         price    high      low       from low\n%s    %s    %s    %s    %s\n"""%(wk_data_1[0],wk_data_1[1],high,low,round((wk_data_1[-3]-low)\/(high-low)*100,2)))\n\n\u8fd0\u884c\u7ed3\u679c\uff1a\n\ntoday         price    high      low       from low\n2019-04-19    30.61    38.939    9.964    70.74<\/code><\/pre>\n
\u6839\u636e\u8fc7\u53bb52\u5468\u6700\u9ad8\u4ef7\u548c\u6700\u4f4e\u4ef7\u7684\u4ea4\u6613\u7b56\u7565<\/p>\n","protected":false},"excerpt":{"rendered":"
\u7edf\u8ba1\u5206\u6790 1\u3001\u8ba1\u7b97\u56de\u62a5\u7387 \u200b \u901a\u5e38\u5728\u91d1\u878d\u884c\u4e1a\u4f1a\u7528\u65e5\u56de\u62a5\u7387\u6765\u8ba1\u7b97\u8d1d\u5854\u503c\uff0c\u4ee5\u8861\u91cf\u4e00\u5bb6\u516c\u53f8\u7684\u5e02\u573a\u98ce\u9669\u3002\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u4ef7 […]<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[317,310],"tags":[],"_links":{"self":[{"href":"https:\/\/egonlin.com\/index.php?rest_route=\/wp\/v2\/posts\/3741"}],"collection":[{"href":"https:\/\/egonlin.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/egonlin.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/egonlin.com\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/egonlin.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3741"}],"version-history":[{"count":0,"href":"https:\/\/egonlin.com\/index.php?rest_route=\/wp\/v2\/posts\/3741\/revisions"}],"wp:attachment":[{"href":"https:\/\/egonlin.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3741"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/egonlin.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3741"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/egonlin.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3741"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}