跳转至

comment

comment_on_rets_and_nets(rets, nets, name='绩效', counts_one_year=12)

输入月频的收益率序列和净值序列，输出年化收益、年化波动、信息比率、月度胜率和最大回撤率 输入2个pd.Series，时间是索引

Parameters

rets : pd.Series 收益率序列，index为时间 nets : pd.Series 净值序列，index为时间 name : str, optional 绩效指标列名字, by default '绩效' counts_one_year : int 一年内有多少次交易, by default 12

Returns

pd.DataFrame 包含年化收益、年化波动、信息比率、月度胜率和最大回撤率的评价指标

Source code in pure_ocean_breeze/labor/comment.py
Python
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
def comment_on_rets_and_nets(
    rets: pd.Series, nets: pd.Series, name: str = "绩效", counts_one_year: int = 12
) -> pd.DataFrame:
    """输入月频的收益率序列和净值序列，输出年化收益、年化波动、信息比率、月度胜率和最大回撤率
    输入2个pd.Series，时间是索引

    Parameters
    ----------
    rets : pd.Series
        收益率序列，index为时间
    nets : pd.Series
        净值序列，index为时间
    name : str, optional
        绩效指标列名字, by default '绩效'
    counts_one_year : int
        一年内有多少次交易, by default 12

    Returns
    -------
    `pd.DataFrame`
        包含年化收益、年化波动、信息比率、月度胜率和最大回撤率的评价指标
    """
    duration_nets = (nets.index[-1] - nets.index[0]).days
    year_nets = duration_nets / 365
    ret_yearly = (nets.iloc[-1] / nets.iloc[0]) ** (1 / year_nets) - 1
    max_draw = ((nets.cummax() - nets) / nets.cummax()).max()
    vol = np.std(rets) * (counts_one_year**0.5)
    info_rate = ret_yearly / vol
    win_rate = len(rets[rets > 0]) / len(rets)
    if counts_one_year == 12:
        names = "月度胜率"
    else:
        names = "胜率"
    comments = pd.DataFrame(
        {
            "年化收益率": ret_yearly,
            "年化波动率": vol,
            "信息比率": info_rate,
            names: win_rate,
            "最大回撤率": max_draw,
        },
        index=[name],
    ).T
    return comments

comments_on_twins(nets, rets, counts_one_year=12)

输入月频的收益率序列和净值序列，给出评价 评价指标包括年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率 输入2个pd.Series，时间是索引

Parameters

nets : pd.Series 净值序列，index为时间 rets : pd.Series 收益率序列，index为时间 counts_one_year : int 一年内有多少次交易, by default 12

Returns

pd.Series 包含年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率的评价指标

Source code in pure_ocean_breeze/labor/comment.py
Python
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
def comments_on_twins(
    nets: pd.Series, rets: pd.Series, counts_one_year: int = 12
) -> pd.Series:
    """输入月频的收益率序列和净值序列，给出评价
    评价指标包括年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率
    输入2个pd.Series，时间是索引

    Parameters
    ----------
    nets : pd.Series
        净值序列，index为时间
    rets : pd.Series
        收益率序列，index为时间
    counts_one_year : int
        一年内有多少次交易, by default 12

    Returns
    -------
    `pd.Series`
        包含年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率的评价指标
    """
    series = nets.copy()
    series1 = rets.copy()
    ret = (series.iloc[-1] - series.iloc[0]) / series.iloc[0]
    duration = (series.index[-1] - series.index[0]).days
    year = duration / 365
    ret_yearly = (series.iloc[-1] / series.iloc[0]) ** (1 / year) - 1
    max_draw = -(series / series.expanding(1).max() - 1).min()
    vol = np.std(series1) * (counts_one_year**0.5)
    sharpe = ret_yearly / vol
    wins = series1[series1 > 0]
    win_rate = len(wins) / len(series1)
    return pd.Series(
        [ret, ret_yearly, vol, sharpe, win_rate, max_draw],
        index=["总收益率", "年化收益率", "年化波动率", "信息比率", "胜率", "最大回撤率"],
    )

comments_on_twins_periods(nets, rets, periods)

输入其他频率的收益率序列和净值序列，给出评价 评价指标包括年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率 输入2个pd.Series，时间是索引

Parameters

nets : pd.Series 净值序列，index为时间 rets : pd.Series 收益率序列，index为时间 periods : int 收益率序列的频率，如5天一次，则为5

Returns

pd.Series 包含年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率的评价指标

Source code in pure_ocean_breeze/labor/comment.py
Python
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
def comments_on_twins_periods(
    nets: pd.Series, rets: pd.Series, periods: int
) -> pd.Series:
    """输入其他频率的收益率序列和净值序列，给出评价
    评价指标包括年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率
    输入2个pd.Series，时间是索引

    Parameters
    ----------
    nets : pd.Series
        净值序列，index为时间
    rets : pd.Series
        收益率序列，index为时间
    periods : int
        收益率序列的频率，如5天一次，则为5

    Returns
    -------
    `pd.Series`
        包含年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率的评价指标
    """
    series = nets.copy()
    series1 = rets.copy()
    ret = (series.iloc[-1] - series.iloc[0]) / series.iloc[0]
    duration = (series.index[-1] - series.index[0]).days
    year = duration / 365
    ret_yearly = (series.iloc[-1] / series.iloc[0]) ** (1 / year) - 1
    max_draw = -(series / series.expanding(1).max() - 1).min()
    vol = np.std(series1) * (252**0.5) * (periods**0.5)
    sharpe = ret_yearly / vol
    wins = series1[series1 > 0]
    win_rate = len(wins) / len(series1)
    return pd.Series(
        [ret, ret_yearly, vol, sharpe, win_rate, max_draw],
        index=["总收益率", "年化收益率", "年化波动率", "信息比率", "胜率", "最大回撤率"],
    )

make_relative_comments(ret_fac, hs300=0, zz500=0, zz1000=0, gz2000=0, all_a=0, day=None, show_nets=0)

对于一个给定的收益率序列，计算其相对于某个指数的超额表现

Parameters

ret_fac : pd.Series 给定的收益率序列，index为时间 hs300 : bool, optional 为1则相对沪深300指数行情, by default 0 zz500 : bool, optional 为1则相对中证500指数行情, by default 0 zz1000 : bool, optional 为1则相对中证1000指数行情, by default 0 gz2000 : bool, optional 为1则相对国证2000指数行情, by default 0 all_a : bool, optional 为1则相对中证全指指数行情, by default 0 day : int, optional 起始日期，形如20130101, by default None show_nets : bool, optional 返回值中包括超额净值数据, by default 0

Returns

pd.Series 评价指标包括年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率

Raises

IOError 如果没指定任何一个指数，将报错

Source code in pure_ocean_breeze/labor/comment.py
Python
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
@do_on_dfs
def make_relative_comments(
    ret_fac: pd.Series,
    hs300: bool = 0,
    zz500: bool = 0,
    zz1000: bool = 0,
    gz2000: bool = 0,
    all_a: bool = 0,
    day: int = None,
    show_nets: bool = 0,
) -> pd.Series:
    """对于一个给定的收益率序列，计算其相对于某个指数的超额表现

    Parameters
    ----------
    ret_fac : pd.Series
        给定的收益率序列，index为时间
    hs300 : bool, optional
        为1则相对沪深300指数行情, by default 0
    zz500 : bool, optional
        为1则相对中证500指数行情, by default 0
    zz1000 : bool, optional
        为1则相对中证1000指数行情, by default 0
    gz2000 : bool, optional
        为1则相对国证2000指数行情, by default 0
    all_a : bool, optional
        为1则相对中证全指指数行情, by default 0
    day : int, optional
        起始日期，形如20130101, by default None
    show_nets : bool, optional
        返回值中包括超额净值数据, by default 0

    Returns
    -------
    `pd.Series`
        评价指标包括年化收益率、总收益率、年化波动率、年化夏普比率、最大回撤率、胜率

    Raises
    ------
    `IOError`
        如果没指定任何一个指数，将报错
    """

    if hs300 == 1 and zz500 == 1 and zz1000 == 0 and gz2000 == 0:
        net_index = read_index_single("000906.SH").resample("M").last()
    else:
        net_indexs = []
        weights = []
        if hs300:
            net_index = read_index_single("000300.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(300)
        if zz500:
            net_index = read_index_single("000905.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(500)
        if zz1000:
            net_index = read_index_single("000852.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(1000)
        if gz2000:
            net_index = read_index_single("399303.SZ").resample("M").last()
            net_indexs.append(net_index)
            weights.append(2000)
        if all_a:
            net_index = read_index_single("000985.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(5000)
        if (hs300 + zz500 + zz1000 + gz2000+ all_a) == 0:
            raise IOError("你总得指定一个股票池吧？")
        net_index = pd.concat(net_indexs, axis=1)
    ret_index = net_index.pct_change()
    if isinstance(ret_index, pd.DataFrame):
        ret_index = sum(
            [ret_index.iloc[:, i] * weights[i] for i in range(len(weights))]
        ) / sum(weights)
    if day is not None:
        ret_index = ret_index[ret_index.index >= pd.Timestamp(day)]
    ret = ret_fac - ret_index
    ret = ret.dropna()
    net = (1 + ret).cumprod()
    ntop = pd.Series(1, index=[net.index.min() - pd.DateOffset(months=1)])
    rtop = pd.Series(0, index=[net.index.min() - pd.DateOffset(months=1)])
    net = pd.concat([ntop, net]).resample("M").last()
    ret = pd.concat([rtop, ret]).resample("M").last()
    com = comments_on_twins(net, ret)
    if show_nets:
        return com, net
    else:
        return com

make_relative_comments_plot(ret_fac, hs300=0, zz500=0, zz1000=0, gz2000=0, all_a=0, day=None)

对于一个给定的收益率序列，计算其相对于某个指数的超额表现，然后绘图，并返回超额净值序列

Parameters

ret_fac : pd.Series 给定的收益率序列，index为时间 hs300 : bool, optional 为1则相对沪深300指数行情, by default 0 zz500 : bool, optional 为1则相对中证500指数行情, by default 0 zz1000 : bool, optional 为1则相对中证1000指数行情, by default 0 gz2000 : bool, optional 为1则相对国证2000指数行情, by default 0 all_a : bool, optional 为1则相对中证全指指数行情, by default 0 day : int, optional 起始日期，形如20130101, by default None

Returns

pd.Series 超额净值序列

Raises

IOError 如果没指定任何一个指数，将报错

Source code in pure_ocean_breeze/labor/comment.py
Python
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
@do_on_dfs
def make_relative_comments_plot(
    ret_fac: pd.Series,
    hs300: bool = 0,
    zz500: bool = 0,
    zz1000: bool = 0,
    gz2000: bool = 0,
    all_a: bool = 0,
    day: int = None,
) -> pd.Series:
    """对于一个给定的收益率序列，计算其相对于某个指数的超额表现，然后绘图，并返回超额净值序列

    Parameters
    ----------
    ret_fac : pd.Series
        给定的收益率序列，index为时间
    hs300 : bool, optional
        为1则相对沪深300指数行情, by default 0
    zz500 : bool, optional
        为1则相对中证500指数行情, by default 0
    zz1000 : bool, optional
        为1则相对中证1000指数行情, by default 0
    gz2000 : bool, optional
        为1则相对国证2000指数行情, by default 0
    all_a : bool, optional
        为1则相对中证全指指数行情, by default 0
    day : int, optional
        起始日期，形如20130101, by default None

    Returns
    -------
    `pd.Series`
        超额净值序列

    Raises
    ------
    `IOError`
        如果没指定任何一个指数，将报错
    """
    if hs300 == 1 and zz500 == 1 and zz1000 == 0 and gz2000 == 0:
        net_index = read_index_single("000906.SH").resample("M").last()
    else:
        net_indexs = []
        weights = []
        if hs300:
            net_index = read_index_single("000300.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(300)
        if zz500:
            net_index = read_index_single("000905.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(500)
        if zz1000:
            net_index = read_index_single("000852.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(1000)
        if gz2000:
            net_index = read_index_single("399303.SZ").resample("M").last()
            net_indexs.append(net_index)
            weights.append(2000)
        if all_a:
            net_index = read_index_single("000985.SH").resample("M").last()
            net_indexs.append(net_index)
            weights.append(5000)
        if (hs300 + zz500 + zz1000 + gz2000+ all_a) == 0:
            raise IOError("你总得指定一个股票池吧？")
        net_index = pd.concat(net_indexs, axis=1)
    ret_index = net_index.pct_change()
    if isinstance(ret_index, pd.DataFrame):
        ret_index = sum(
            [ret_index.iloc[:, i] * weights[i] for i in range(len(weights))]
        ) / sum(weights)
    if day is not None:
        ret_index = ret_index[ret_index.index >= pd.Timestamp(day)]
    ret = ret_fac - ret_index
    ret = ret.dropna()
    net = (1 + ret).cumprod()
    ntop = pd.Series(1, index=[net.index.min() - pd.DateOffset(months=1)])
    net = pd.concat([ntop, net]).resample("M").last()
    net.plot(rot=60)
    plt.show()
    return net

other_periods_comments_nets(fac, way, period, comments_writer=None, nets_writer=None, sheetname=None, group_num=10)

小型回测框架，不同频率下的评价指标，请输入行业市值中性化后的因子值

Parameters

fac : pd.DataFrame 行业市值中性化之后的因子值，index为时间，columns为股票代码 way : str 因子的方向，可选值 pos 或 neg period : int 频率，如5天则为5, by default None comments_writer : pd.ExcelWriter, optional 写入绩效的xlsx, by default None nets_writer : pd.ExcelWriter, optional 写入净值的xlsx, by default None sheetname : str, optional 工作表名称, by default None group_num : int, optional 回测时分组数量, by default 10

Returns

Tuple[pd.Series] 绩效和净值

Source code in pure_ocean_breeze/labor/comment.py
Python
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
def other_periods_comments_nets(
    fac: pd.DataFrame,
    way: str,
    period: int,
    comments_writer: pd.ExcelWriter = None,
    nets_writer: pd.ExcelWriter = None,
    sheetname: str = None,
    group_num: int = 10,
) -> Tuple[pd.Series]:
    """小型回测框架，不同频率下的评价指标，请输入行业市值中性化后的因子值

    Parameters
    ----------
    fac : pd.DataFrame
        行业市值中性化之后的因子值，index为时间，columns为股票代码
    way : str
        因子的方向，可选值 pos 或 neg
    period : int
        频率，如5天则为5, by default None
    comments_writer : pd.ExcelWriter, optional
        写入绩效的xlsx, by default None
    nets_writer : pd.ExcelWriter, optional
        写入净值的xlsx, by default None
    sheetname : str, optional
        工作表名称, by default None
    group_num : int, optional
        回测时分组数量, by default 10

    Returns
    -------
    `Tuple[pd.Series]`
        绩效和净值
    """
    import alphalens as al

    closes = read_daily(open=1).shift(-1)
    fac1 = fac.stack()
    df = al.utils.get_clean_factor_and_forward_returns(
        fac1,
        closes[closes.index.isin(fac.index)],
        quantiles=group_num,
        periods=(period,),
    )
    df = df.reset_index()
    ics = df.groupby(["date"])[[f"{period}D", "factor"]].apply(
        lambda x: x.corr(method="spearman").iloc[0, 1]
    )
    ic = ics.mean()
    ir = ics.std()
    icir = ic / ir * (252**0.5) / (period**0.5)
    df = df.groupby(["date", "factor_quantile"])[f"{period}D"].mean() / period
    df = df.unstack()
    df.columns = [f"分组{i}" for i in list(df.columns)]
    if way == "pos":
        df = df.assign(多空对冲=df[f"分组{group_num}"] - df.分组1)
    elif way == "neg":
        df = df.assign(多空对冲=df.分组1 - df[f"分组{group_num}"])
    nets = (df + 1).cumprod()
    nets = nets.apply(lambda x: x / x.iloc[0])
    nets.plot(rot=60)
    plt.show()
    comments = comments_on_twins_periods(nets.多空对冲, df.多空对冲, period)
    comments = pd.concat(
        [pd.Series([ic, icir], index=["Rank IC", "Rank ICIR"]), comments]
    )
    print(comments)
    if sheetname is None:
        ...
    else:
        if comments_writer is None:
            ...
        else:
            comments.to_excel(comments_writer, sheetname)
        if nets_writer is None:
            ...
        else:
            nets.to_excel(nets_writer, sheetname)
    return comments, nets