>>> import valuequant as vq
>>> vq.login(name="BoomEvolve账户名称",pswd="账户密码")
登录成功
                            

>>> #构建线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createlinear()

>>> #构建严格状态更新模型
>>> model=vq.Models.strict_latest_status_model(series=series,minstep=8,significance=0.3)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 2.000001, 'value_latest': 90.000137}, 'status': {'l1': '+', 'l2': 'n'}}

>>> #构建含季节项的线性增长测试序列
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的严格状态更新模型
>>> model=vq.Models.strict_latest_status_model(series=series,minstep=8,significance=0.3,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 2.000039, 'value_latest': 90.002251}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.997975, '3': 1.000952, '6': 1.002025, '9': 0.999045}, 'qorder': ['3', '6', '9', '12']}

>>> #构建二次增长测试序列
>>> def createquadratic(q=-0.08,l=10,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
series=createquadratic()

>>> #构建严格状态更新模型
>>> model=vq.Models.strict_latest_status_model(series=series,minstep=8,significance=0.3)
>>> model.model()
{'method': 'l2', 'step': 40, 'params': {'l2_diff': -0.159991, 'l1_latest': 3.6801569999999515, 'value_latest': 282.000145}, 'status': {'l1': '+', 'l2': '-'}}
>>> #默认conservative参数为False，若1阶差分状态和2阶差分状态不同，曲线方向将在未来有限序列步出现反转

>>> #构建保守的严格状态更新模型
>>> model=vq.Models.strict_latest_status_model(series=series,minstep=8,significance=0.3,conservative=True)
>>> model.model()
{'method': 'l2smooth', 'step': 40, 'params': {'l1_latest': 3.6801569999999515, 'value_latest': 282.000145, 'l2smooth_rate': 0.9746603487603882}, 'status': {'l1': '+', 'l2': '-'}, 'smooth': True}
>>> #设置conservative参数为True，则在分析者对时间序列反转缺乏信心的情况下提供保守的估计方式，尽量使得未来推演在不出现反转的情况下实现1阶差分绝对值逐渐缩小
                            

>>> #构建线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createlinear()

>>> #构建概率状态更新模型
>>> model=vq.Models.probable_latest_status_model(series=series,minstep=8,significance=0.3)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 26, 'params': {'l1_diff': 2.0, 'value_latest': 90.000023}, 'status': {'l1': '+', 'l2': 'n'}}

>>> #构建含季节项的线性增长测试序列
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的概率状态更新模型
>>> model=vq.Models.probable_latest_status_model(series=series,minstep=8,significance=0.3,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 26, 'params': {'l1_diff': 2.000159, 'value_latest': 90.002191}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.997975, '3': 1.000952, '6': 1.002024, '9': 0.999045}, 'qorder': ['3', '6', '9', '12']}

>>> #构建二次增长测试序列
>>> def createquadratic(q=-0.08,l=10,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
series=createquadratic()

>>> #构建概率状态更新模型
>>> model=vq.Models.probable_latest_status_model(series=series,minstep=8,significance=0.3)
>>> model.model()
{'method': 'l2', 'step': 26, 'params': {'l2_diff': -0.159994, 'l1_latest': 3.6799239999999713, 'value_latest': 282.000051}, 'status': {'l1': '+', 'l2': '-'}}
>>> #默认conservative参数为False，若1阶差分状态和2阶差分状态不同，曲线方向将在未来有限序列步出现反转

>>> #构建保守的概率状态更新模型
>>> model=vq.Models.probable_latest_status_model(series=series,minstep=8,significance=0.3,conservative=True)
>>> model.model()
{'method': 'l2smooth', 'step': 26, 'params': {'l1_latest': 3.6799239999999713, 'value_latest': 282.000051, 'l2smooth_rate': 0.9706620971665002}, 'status': {'l1': '+', 'l2': '-'}, 'smooth': True}
>>> #设置conservative参数为True，则在分析者对时间序列反转缺乏信心的情况下提供保守的估计方式，尽量使得未来推演在不出现反转的情况下实现1阶差分绝对值逐渐缩小
                            

>>> #构建线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createlinear()

>>> #构建长期状态模型
>>> model=vq.Models.long_range_status_model(series=series,significance=0.3)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 2.000003, 'value_latest': 90.000072}, 'status': {'l1': '+', 'l2': 'n'}}

>>> #构建含季节项的线性增长测试序列
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的长期状态模型
>>> model=vq.Models.long_range_status_model(series=series,significance=0.3,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 2.000045, 'value_latest': 90.002203}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.997976, '3': 1.000953, '6': 1.002024, '9': 0.999043}, 'qorder': ['3', '6', '9', '12']}

>>> #构建二次增长测试序列
>>> def createquadratic(q=-0.08,l=10,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
series=createquadratic()

>>> #构建长期状态模型
>>> model=vq.Models.long_range_status_model(series=series,significance=0.3)
>>> model.model()
{'method': 'l2', 'step': 40, 'params': {'l2_diff': -0.159992, 'l1_latest': 3.6800779999999804, 'value_latest': 282.00007}, 'status': {'l1': '+', 'l2': '-'}}
>>> #默认conservative参数为False，若1阶差分状态和2阶差分状态不同，曲线方向将在未来有限序列步出现反转

>>> #构建保守的长期状态模型
>>> model=vq.Models.long_range_status_model(series=series,significance=0.3,conservative=True)
>>> model.model()
{'method': 'l2smooth', 'step': 40, 'params': {'l1_latest': 3.6800779999999804, 'value_latest': 282.00007, 'l2smooth_rate': 0.9746599006548765}, 'status': {'l1': '+', 'l2': '-'}, 'smooth': True}
>>> #设置conservative参数为True，则在分析者对时间序列反转缺乏信心的情况下提供保守的估计方式，尽量使得未来推演在不出现反转的情况下实现1阶差分绝对值逐渐缩小
                            

>>> #构建随机-线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createrandom(b=10,sigma=2,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> def createlinear(l=1,b=10,sigma=0.0001,startdate='2019-03-31',freq='Q-DEC',periods=12):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createrandom()
>>> series['2019':]=createlinear()

>>> #构建严格状态长期惯性模型
>>> model=vq.Models.strict_latest_long_range_model(series,minstep=8,significance=0.3,inertia=3)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'params': {'l1_diff': 0.999982, 'value_latest': 21.999832}, 'strict_latest': {'method': 'l1', 'step': 13, 'params': {'l1_diff': 0.999982, 'value_latest': 21.999832}, 'status': {'l1': '+', 'l2': 'n'}}, 'long_range': {'method': 'random', 'step': 40, 'params': {'value_mean': 11.416678}, 'status': {'l1': 'n'}}, 'inertia': 3, 'change': True, 'change_method': 'random', 'change_status': {'l1': 'n'}}
                            

>>> #构建随机-线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createrandom(b=10,sigma=2,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> def createlinear(l=1,b=10,sigma=0.0001,startdate='2019-03-31',freq='Q-DEC',periods=12):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createrandom()
>>> series['2019':]=createlinear()

>>> #构建概率状态长期惯性模型
>>> model=vq.Models.probable_latest_long_range_model(series,minstep=8,significance=0.3,inertia=3)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'params': {'l1_diff': 0.999993, 'value_latest': 21.999974}, 'probable_latest': {'method': 'l1', 'step': 9, 'params': {'l1_diff': 0.999993, 'value_latest': 21.999974}, 'status': {'l1': '+', 'l2': 'n'}}, 'long_range': {'method': 'random', 'step': 40, 'params': {'value_mean': 11.579264}, 'status': {'l1': 'n'}}, 'inertia': 3, 'change': True, 'change_method': 'random', 'change_status': {'l1': 'n'}}
                            

>>> #构建线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createlinear()

>>> #构建移动平均严格状态更新模型
>>> model=vq.Models.ma_strict_latest_status_model(series,mawindow=4,minstep=8)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 1.999999, 'value_latest': 89.999943}, 'status': {'l1': '+', 'l2': 'n'}}

>>> #构建含季节项的线性增长测试序列
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的移动平均严格状态更新模型
>>> model=vq.Models.ma_strict_latest_status_model(series,mawindow=4,minstep=8,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 2.0, 'value_latest': 90.00241}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.997973, '3': 1.000953, '6': 1.002024, '9': 0.999045}, 'qorder': ['3', '6', '9', '12']}

>>> #构建含季节项的二次增长测试序列
>>> def createquadratic(q=-0.08,l=10,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createquadratic()*createseasonal()

>>> #构建移动平均严格状态更新模型
>>> model=vq.Models.ma_strict_latest_status_model(series,mawindow=4,minstep=8,seasonal=True)
>>> model.model()
{'method': 'l2', 'step': 40, 'params': {'l2_diff': -0.159866, 'l1_latest': 3.657406999999978, 'value_latest': 281.744325}, 'status': {'l1': '+', 'l2': '-'}, 'seasonal': {'12': 0.998906, '3': 1.001805, '6': 1.00287, '9': 0.999837}, 'qorder': ['3', '6', '9', '12']}
>>> #默认conservative参数为False，若1阶差分状态和2阶差分状态不同，曲线方向将在未来有限序列步出现反转

>>> #构建保守的移动平均严格状态更新模型
>>> model=vq.Models.ma_strict_latest_status_model(series,mawindow=4,minstep=8,seasonal=True,conservative=True)
>>> model.model()
{'method': 'l2smooth', 'step': 40, 'params': {'l1_latest': 3.657406999999978, 'value_latest': 281.744325, 'l2smooth_rate': 0.9745267582750817}, 'status': {'l1': '+', 'l2': '-'}, 'seasonal': {'12': 0.998906, '3': 1.001805, '6': 1.00287, '9': 0.999837}, 'qorder': ['3', '6', '9', '12'], 'smooth': True}
>>> #设置conservative参数为True，则在分析者对时间序列反转缺乏信心的情况下提供保守的估计方式，尽量使得未来推演在不出现反转的情况下实现1阶差分绝对值逐渐缩小
                            

>>> #构建含季节项的线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的移动平均概率状态更新模型
>>> model=vq.Models.ma_probable_latest_status_model(series,mawindow=4,minstep=8,significance=0.3,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 41, 'params': {'l1_diff': 2.000006, 'value_latest': 90.002236}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.997975, '3': 1.000952, '6': 1.002024, '9': 0.999045}, 'qorder': ['3', '6', '9', '12']}

>>> #构建含季节项的二次增长测试序列
>>> def createquadratic(q=-0.08,l=10,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createquadratic()*createseasonal()

>>> #构建保守的移动平均概率状态更新模型
>>> model=vq.Models.ma_probable_latest_status_model(series,mawindow=4,minstep=8,significance=0.3,seasonal=True,conservative=True)
>>> model.model()
{'method': 'l2smooth', 'step': 41, 'params': {'l1_latest': 3.657123000000013, 'value_latest': 281.744139, 'l2smooth_rate': 0.9745239619673073}, 'status': {'l1': '+', 'l2': '-'}, 'seasonal': {'12': 0.998906, '3': 1.001805, '6': 1.002871, '9': 0.999837}, 'qorder': ['3', '6', '9', '12'], 'smooth': True}
>>> #设置conservative参数为True，则在分析者对时间序列反转缺乏信心的情况下提供保守的估计方式，尽量使得未来推演在不出现反转的情况下实现1阶差分绝对值逐渐缩小
                            

>>> #构建含季节项的线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的移动平均长期状态模型
>>> model=model=vq.Models.ma_long_range_status_model(series,mawindow=4,significance=0.3,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 2.0, 'value_latest': 90.002131}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.997975, '3': 1.000953, '6': 1.002024, '9': 0.999043}, 'qorder': ['3', '6', '9', '12']}

>>> #构建含季节项的二次增长测试序列
>>> def createquadratic(q=-0.08,l=10,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createquadratic()*createseasonal()

>>> #构建保守的移动平均长期状态模型
>>> model=vq.Models.ma_long_range_status_model(series,mawindow=4,significance=0.3,seasonal=True,conservative=True)
>>> model.model()
{'method': 'l2smooth', 'step': 40, 'params': {'l1_latest': 3.6572810000000118, 'value_latest': 281.744331, 'l2smooth_rate': 0.9745257669178147}, 'status': {'l1': '+', 'l2': '-'}, 'seasonal': {'12': 0.998906, '3': 1.001805, '6': 1.002871, '9': 0.999837}, 'qorder': ['3', '6', '9', '12'], 'smooth': True}
>>> #设置conservative参数为True，则在分析者对时间序列反转缺乏信心的情况下提供保守的估计方式，尽量使得未来推演在不出现反转的情况下实现1阶差分绝对值逐渐缩小
                            

>>> #构建随机测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createrandom(b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createrandom()

>>> #构建序列形态模型
>>> model=vq.Models.series_shape_model(series=series,significance=0.3)
>>> #获取建模结果
>>> model.model()
{'params': {'constant': 10.00001}, 'sigma': 9.5e-05, 'method': 'random', 'status': ['n', 'n', 'n']}

>>> #构建线性增长测试序列
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> series=createlinear()

>>> #构建序列形态模型
>>> model=vq.Models.series_shape_model(series=series,significance=0.3)
>>> #获取建模结果
>>> model.model()
{'params': {'linear': 2.0, 'constant': 90.000008}, 'sigma': 0.000106, 'r2': 1.0, 'method': 'linear', 'status': ['+', 'n', 'n']}

>>> #构建二次增长测试序列
>>> def createquadratic(q=-0.08,l=10,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createquadratic()

>>> #构建序列形态模型
>>> model=vq.Models.series_shape_model(series=series,significance=0.3)
>>> #获取建模结果
>>> model.model()
{'params': {'quadratic': -0.08, 'linear': 3.600005, 'constant': 282.000055}, 'sigma': 0.000109, 'r2': 1.0, 'method': 'quadratic', 'status': ['+', '-', 'n']}

>>> #构建三次增长测试序列
>>> def createcubic(c=1,q=-2,l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(np.arange(1, periods + 1),3)*c+pow(np.arange(1, periods + 1),2)*q+np.arange(1, periods + 1) * l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createcubic()

>>> #构建序列形态模型
>>> model=vq.Models.series_shape_model(series=series,significance=0.3)
>>> #获取建模结果
>>> model.model()
{'params': {'cubic': 1.0, 'quadratic': 118.0, 'linear': 4641.999996, 'constant': 60890.000047}, 'sigma': 9.3e-05, 'r2': 1.0, 'method': 'cubic', 'status': ['+', '+', '+']}

>>> #构建指数增长测试序列
>>> def createexponential(l=-6,p=0.9,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(pow(p,np.arange(1, periods + 1))*l + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createexponential()

>>> #构建序列形态模型
>>> model=vq.Models.series_shape_model(series=series,significance=0.3,conservative=True)
>>> #获取建模结果
>>> model.model()
{'params': {'powerbase': 0.900004, 'linear': -0.088699, 'constant': 10.00004}, 'sigma': 8.7e-05, 'r2': 1.0, 'method': 'exponential', 'status': ['+', '-', '+']}

>>> #构建逻辑斯蒂增长测试序列
>>> def createlogistic(K=20,a=10,g=0.1,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(K/(1+a*np.exp(-g*np.arange(1, periods + 1))) + b + np.random.normal(0, sigma, periods), index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlogistic()

>>> #构建序列形态模型
>>> model=vq.Models.series_shape_model(series=series,significance=0.3,conservative=True)
>>> #获取建模结果
>>> model.model()
{'params': {'yq': -0.005048, 'yl': 0.200283, 'yb': -1.486114, 'vl': 26.903902}, 'sigma': 0.001704, 'r2': 0.999692, 'method': 'logistic', 'status': ['+', 'n', '-']}
                            

>>> #构建含季节项的线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的移动平均严格更新序列形态模型
>>> model=vq.Models.ma_strict_latest_status_shape_model(series,mawindow=4,minstep=8,significance=0.3,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'params': {'linear': 2.000002, 'constant': 90.000027}, 'sigma': 0.002126, 'r2': 1.0, 'method': 'linear', 'status': ['+', 'n', 'n'], 'step': 39, 'seasonal': {'12': 0.997976, '3': 1.000953, '6': 1.002022, '9': 0.999045}, 'qorder': ['3', '6', '9', '12']}
                            

>>> #构建含季节项的线性增长测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> def createseasonal(seasonal=[1.001,1.002,0.999,0.998],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建设置季节项的移动平均概率更新序列形态模型
>>> model=vq.Models.ma_probable_latest_status_shape_model(series,mawindow=4,minstep=8,significance=0.3,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'params': {'linear': 1.999999, 'constant': 89.999998}, 'sigma': 0.0021, 'r2': 1.0, 'method': 'linear', 'status': ['+', 'n', 'n'], 'step': 41, 'seasonal': {'12': 0.997976, '3': 1.000953, '6': 1.002022, '9': 0.999045}, 'qorder': ['3', '6', '9', '12']}
                            

>>> #构建测试序列
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> def createseasonal(seasonal=[1.01,1.02,0.99,0.98],startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.tile(seasonal,int(np.ceil(periods/len(seasonal))))[:periods], index=pd.date_range(start=startdate, freq=freq, periods=periods))
>>> series=createlinear()*createseasonal()

>>> #构建模型
>>> model=vq.Models.ma_strict_latest_status_model(series,mawindow=4,minstep=8,seasonal=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 40, 'params': {'l1_diff': 1.999994, 'value_latest': 90.02255}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.979754, '3': 1.009525, '6': 1.020241, '9': 0.990442}, 'qorder': ['3', '6', '9', '12']}

>>> #查询样本数据
>>> model.data()
2012-03-31    12.119795
2012-06-30    14.280024
2012-09-30    15.839959
2012-12-31    17.639740
......
2021-03-31    84.839858
2021-06-30    87.720110
2021-09-30    87.120009
2021-12-31    88.199922
Freq: Q-DEC, dtype: float64

>>> #模型样本内拟合
>>> model.fit()
2012-03-31          NaN
2012-06-30    14.405588
2012-09-30    16.124414
2012-12-31    17.478765
......
2021-03-31    83.144534
2021-06-30    88.597577
2021-09-30    88.862559
2021-12-31    87.315679
Freq: Q-DEC, dtype: float64

>>> #模型样本外推演
>>> model.forecast(t=20)
2022-03-31     92.899059
2022-06-30     95.925648
2022-09-30     95.104749
2022-12-31     96.037962
......
2026-03-31    125.203762
2026-06-30    128.573262
2026-09-30    126.798798
2026-12-31    127.389996
Freq: Q-DEC, dtype: float64

>>> #结构化模型的的现值计算
>>> model.pv(r=0.1)
7432.00662

>>> #结构化模型的交易日价格内部收益率计算
>>> model.irr(price=7432.00662,tradedate='2022-03-31')
0.1

>>> #构建测试交易日价格序列
>>> priceseries=pd.Series([500,1000,5000,10000,50000],index=pd.date_range(start='2022-01-01',periods=5))
>>> #结构化模型的交易日价格序列内部收益率计算
>>> model.batchirr(priceseries=priceseries)
                IRRs errors
2022-01-01       NaN   over
2022-01-02  0.520891   None
2022-01-03  0.130965   None
2022-01-04  0.080250   None
2022-01-05  0.029575   None

>>> #结构化模型的交易日价格兑现率计算
>>> model.achieverate(price=7432.00662,tradedate='2022-03-31',r=0.1)
1.0

>>> #结构化模型的交易日价格序列兑现率计算
>>> model.batchachieverate(priceseries=priceseries,r=0.1)
2022-01-01    14.517870
2022-01-02     7.260857
2022-01-03     1.452556
2022-01-04     0.726470
2022-01-05     0.145333
Freq: D, dtype: float64
                            

>>> #构建测试时间序列集数据表和测试时间序列集结构
>>> import numpy as np
>>> import pandas as pd
>>> def createlinear(l=2,b=10,sigma=0.0001,startdate='2012-03-31',freq='Q-DEC',periods=40):
...     return pd.Series(np.arange(1,periods+1)*l+b+np.random.normal(0,sigma,periods),index=pd.date_range(start=startdate,freq=freq,periods=periods))
>>> df=pd.DataFrame()
>>> df['s1']=createlinear()
>>> df['s2']=df['s1']*createlinear(l=0.01, b=0.1, sigma=0.0001)
>>> df['s3']=createlinear(l=1, b=5)
>>> df['s4']=df['s1']-df['s3']
>>> df['s5']=0.5*df['s4']
>>> structure={
...     's1':{'mode':'auto'},
...     's2':{'mode':'derive','base':'s1'},
...     's3':{'mode':'auto'},
...     's4':{'mode': 'compo', 'plus': ['s1'], 'minus': ['s3']},
...     's5':{'mode': 'quota', 'quota':0.5 ,'base':'s4'}
... }

>>> #构建模型集
>>> modelset=vq.ModelSets.series_shape_modelset(df=df,structure=structure,significance=0.3)
>>> #获取建模结果
>>> modelset.modelset()
{'s1': {'mode': 'auto', 'model': {'params': {'linear': 1.999999, 'constant': 89.999985}, 'sigma': 0.0001, 'r2': 1.0, 'method': 'linear', 'status': ['+', 'n', 'n']}}, 's2': {'mode': 'derive', 'base': 's1', 'model': {'params': {'linear': 0.009999, 'constant': 0.499966}, 'sigma': 0.000104, 'r2': 0.999999, 'method': 'linear', 'status': ['+', 'n', 'n']}}, 's3': {'mode': 'auto', 'model': {'params': {'linear': 1.0, 'constant': 44.999949}, 'sigma': 0.000104, 'r2': 1.0, 'method': 'linear', 'status': ['+', 'n', 'n']}}, 's4': {'mode': 'compo', 'plus': ['s1'], 'minus': ['s3']}, 's5': {'mode': 'quota', 'quota': 0.5, 'base': 's4'}}

>>> #查询维度样本数据
>>> modelset.data('s1')
2012-03-31    12.000140
2012-06-30    13.999952
2012-09-30    15.999933
2012-12-31    18.000088
......
2021-03-31    83.999815
2021-06-30    85.999978
2021-09-30    88.000075
2021-12-31    89.999987
Freq: Q-DEC, Name: s1, dtype: float64

>>> #维度模型样本内拟合
>>> modelset.fit('s1')
2012-03-31    12.000024
2012-06-30    14.000023
2012-09-30    16.000022
2012-12-31    18.000021
......
2021-03-31    83.999988
2021-06-30    85.999987
2021-09-30    87.999986
2021-12-31    89.999985
Freq: Q-DEC, dtype: float64

>>> #维度模型样本外推演
>>> modelset.forecast('s1',t=20)
2022-03-31     91.999984
2022-06-30     93.999983
2022-09-30     95.999982
2022-12-31     97.999981
......
2026-03-31    123.999968
2026-06-30    125.999967
2026-09-30    127.999966
2026-12-31    129.999965
Freq: Q-DEC, dtype: float64

>>> #结构化模型集维度模型的现值计算
>>> modelset.pv('s1',r=0.025)
7051.997704

>>> #结构化模型集维度模型的交易日价格内部收益率计算
>>> modelset.irr('s1',price=7051.997704,tradedate='2022-03-31')
0.025

>>> #构建测试交易日价格序列
>>> priceseries=pd.Series([500,1000,5000,10000,50000],index=pd.date_range(start='2022-01-01',periods=5))
>>> #结构化模型集维度模型的交易日价格序列内部收益率计算
>>> modelset.batchirr('s1',priceseries=priceseries)
                IRRs errors
2022-01-01  0.204026   None
2022-01-02  0.110376   None
2022-01-03  0.031237   None
2022-01-04  0.019483   None
2022-01-05  0.007313   None

>>> #结构化模型集维度模型的交易日价格兑现率计算
>>> modelset.achieverate('s1',price=7051.997704,tradedate='2022-03-31',r=0.025)
1.0

>>> #结构化模型集维度模型的交易日价格序列兑现率计算
>>> modelset.batchachieverate('s1',priceseries=priceseries,r=0.025)
2022-01-01    13.763771
2022-01-02     6.883774
2022-01-03     1.377133
2022-01-04     0.688755
2022-01-05     0.137789
Freq: D, dtype: float64
                            

>>> #构建模型
>>> model=vq.StockModels.ma_strict_latest_status_model(stockcode='600036',dimension='利息收入',reportdate='2021-09-30',mawindow=4,minstep=8,seasonal=True,conservative=True)
>>> #获取建模结果
>>> model.model()
{'method': 'l1', 'step': 24, 'params': {'l1_diff': 0.068282, 'value_latest': 3.261846}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 1.001776, '3': 0.98824, '6': 0.993877, '9': 1.00486}, 'qorder': ['12', '3', '6', '9']}

>>> #查询证券利润表维度时间序列数据
>>> model.incdata()
2005-03-31    0.156246
2005-06-30    0.000000
2005-09-30    0.000000
2005-12-31    0.380716
......
2020-12-31    2.964659
2021-03-31    3.054092
2021-06-30    3.103159
2021-09-30    3.277697
Length: 67, dtype: float64

>>> #模型样本内拟合
>>> model.fit()
2015-12-31    1.995791
2016-03-31    1.945666
2016-06-30    1.893261
2016-09-30    1.818240
......
2020-12-31    3.053510
2021-03-31    2.997274
2021-06-30    3.103256
2021-09-30    3.186854
Freq: Q-DEC, dtype: float64

>>> #模型样本外推演
>>> model.forecast(t=20)
2021-12-31    3.336042
2022-03-31    3.358445
2022-06-30    3.445465
2022-09-30    3.552154
......
2025-12-31    4.430495
2026-03-31    4.438109
2026-06-30    4.531288
2026-09-30    4.649976
Freq: Q-DEC, dtype: float64

>>> #现值计算
>>> model.pv(r=0.1)
260.950909

>>> #交易日价格内部收益率计算
>>> model.irr(price=260.950909,tradedate='2021-12-31')
0.1

>>> #构建测试交易日价格序列
>>> import pandas as pd
>>> import numpy as np
>>> priceseries=pd.Series([500,1000,5000,10000,50000],index=pd.date_range(start='2022-01-01',periods=5))
>>> #交易日价格序列内部收益率计算
>>> model.batchirr(priceseries=priceseries)
                IRRs errors
2022-01-01  0.063911   None
2022-01-02  0.041181   None
2022-01-03  0.016284   None
2022-01-04  0.011187   None
2022-01-05       NaN  under

>>> #交易日价格兑现率计算
>>> model.achieverate(price=260.950909,tradedate='2021-12-31',r=0.1)
1.0

>>> #交易日价格序列兑现率计算
>>> model.batchachieverate(priceseries=priceseries,r=0.1)
2022-01-01    0.522040
2022-01-02    0.261089
2022-01-03    0.052232
2022-01-04    0.026123
2022-01-05    0.005226
Freq: D, dtype: float64

>>> #实际交易日价格内部收益率计算
>>> model.irr_trade(pricemode='close',tradedate='2021-11-08')
0.377432

>>> #实际交易时期价格内部收益率计算
>>> model.batchirr_trade(pricemode='close',startdate='2021-10-08',enddate='2021-10-18')
                IRRs errors
2021-10-08  0.370287   None
2021-10-11  0.357606   None
2021-10-12  0.357323   None
2021-10-13  0.357985   None
2021-10-14  0.362560   None
2021-10-15  0.356793   None
2021-10-18  0.367036   None

>>> #实际交易日价格兑现率计算
>>> model.achieverate_trade(pricemode='close',tradedate='2021-11-08',r=0.1)
5.015884

>>> #实际交易日价格序列兑现率计算
>>> model.batchirr_trade(pricemode='close',startdate='2021-10-08',enddate='2021-10-18')
                IRRs errors
2021-10-08  0.370287   None
2021-10-11  0.357606   None
2021-10-12  0.357323   None
2021-10-13  0.357985   None
2021-10-14  0.362560   None
2021-10-15  0.356793   None
2021-10-18  0.367036   None

>>> #使用TTM年度数据建模
>>> model=vq.StockModels.ma_strict_latest_status_model(stockcode='600036',dimension='利息收入',reportdate='2021-09-30',mawindow=4,minstep=8,conservative=True,ttm=True)
>>> model.model()
{'method': 'l1', 'step': 15, 'params': {'l1_diff': 0.812402, 'value_latest': 12.399607}, 'status': {'l1': '+', 'l2': 'n'}}
>>> #设置参数ttm=True时，seasonal参数无意义
                            

>>> #构建时间序列集结构
>>> structure={
...             '营业收入': {'mode':'auto'},
...             '营业总收入其他': {'mode':'auto'},
...             '营业其他损益': {'mode':'auto'},
...             '营业外收入': {'mode':'auto'},
...             '营业外支出': {'mode':'auto'},
...             '利润总额其他': {'mode':'auto'},
...             '净利润其他': {'mode':'auto'},
...             '综合收益其他': {'mode':'auto'},
...             '营业成本': {'mode':'derive','base':'营业收入'},
...             '营业税金及附加': {'mode':'derive','base':'营业收入'},
...             '销售费用': {'mode':'derive','base':'营业收入'},
...             '管理费用': {'mode':'derive','base':'营业收入'},
...             '财务费用': {'mode':'derive','base':'营业收入'},
...             '营业其他费用': {'mode':'derive','base':'营业收入'},
...             '营业总收入' : {'mode': 'compo', 'plus': ['营业收入','营业总收入其他'], 'minus': []},
...             '营业总成本': {'mode': 'compo', 'plus': ['营业成本','营业税金及附加','销售费用','管理费用','财务费用','营业其他费用'], 'minus': []},
...             '营业利润': {'mode': 'compo', 'plus': ['营业总收入','营业其他损益'], 'minus': ['营业总成本']},
...             '利润总额': {'mode': 'compo', 'plus': ['营业利润','营业外收入','利润总额其他'], 'minus': ['营业外支出']},
...             '税后利润': {'mode': 'quota', 'quota':0.75 ,'base':'利润总额'},
...             '净利润': {'mode': 'compo', 'plus': ['税后利润','净利润其他'], 'minus': []},
...             '综合收益总额':{'mode': 'compo', 'plus': ['净利润','综合收益其他'], 'minus': []},
...         }

>>> #构建模型集
>>> modelset=vq.StockModelSets.ma_strict_latest_status_modelset(stockcode='600690',structure=structure,reportdate='2017-12-31',mawindow=4,minstep=8,seasonal=True,conservative=True)
>>> #获取建模结果
>>> modelset.modelset()
{'营业收入': {'mode': 'auto', 'model': {'method': 'l1', 'step': 14, 'params': {'l1_diff': 0.379913, 'value_latest': 7.85605}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.836401, '3': 0.963235, '6': 1.190104, '9': 0.977119}, 'qorder': ['3', '6', '9', '12']}}, '营业外收入': {'mode': 'auto', 'model': {'method': 'random', 'step': 32, 'params': {'value_mean': 0.024357}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.145846, '3': 0.769503, '6': 0.789491, '9': 0.952866}, 'qorder': ['3', '6', '9', '12']}}, '营业外支出': {'mode': 'auto', 'model': {'method': 'random', 'step': 58, 'params': {'value_mean': 0.001818}, 'status': {'l1': 'n'}, 'seasonal': {'12': 0.848634, '3': 0.602179, '6': 1.029571, '9': 1.205745}, 'qorder': ['3', '6', '9', '12']}}, '综合收益其他': {'mode': 'auto', 'model': {'method': 'random', 'step': 32, 'params': {'value_mean': -0.001163}, 'status': {'l1': 'n'}, 'seasonal': {'12': -0.094787, '3': 1.698016, '6': 1.446444, '9': 2.681057}, 'qorder': ['3', '6', '9', '12']}}, '营业成本': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 14, 'params': {'value_mean': 0.699182}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.002891, '3': 0.994847, '6': 0.988821, '9': 1.012805}, 'qorder': ['3', '6', '9', '12']}}, '营业税金及附加': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 58, 'params': {'value_mean': 0.003632}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.035196, '3': 0.840581, '6': 0.928841, '9': 1.04492}, 'qorder': ['3', '6', '9', '12']}}, '销售费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 45, 'params': {'value_mean': 0.137667}, 'status': {'l1': 'n'}, 'seasonal': {'12': 0.951441, '3': 1.088444, '6': 1.018624, '9': 0.943284}, 'qorder': ['3', '6', '9', '12']}}, '管理费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 23, 'params': {'value_mean': 0.068995}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.087347, '3': 1.017571, '6': 0.944272, '9': 0.942817}, 'qorder': ['3', '6', '9', '12']}}, '财务费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 58, 'params': {'value_mean': 0.000483}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.236724, '3': 1.338647, '6': 1.909793, '9': 0.460659}, 'qorder': ['3', '6', '9', '12']}}, '营业其他费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 48, 'params': {'value_mean': 0.001753}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.93244, '3': 1.349546, '6': 0.171696, '9': 0.833086}, 'qorder': ['3', '6', '9', '12']}}, '营业总收入': {'mode': 'compo', 'plus': ['营业收入'], 'minus': []}, '营业总成本': {'mode': 'compo', 'plus': ['营业成本', '营业税金及附加', '销售费用', '管理费用', '财务费用', '营业其他费用'], 'minus': []}, '营业利润': {'mode': 'compo', 'plus': ['营业总收入'], 'minus': ['营业总成本']}, '利润总额': {'mode': 'compo', 'plus': ['营业利润', '营业外收入'], 'minus': ['营业外支出']}, '税后利润': {'mode': 'quota', 'quota': 0.75, 'base': '利润总额'}, '净利润': {'mode': 'compo', 'plus': ['税后利润'], 'minus': []}, '综合收益总额': {'mode': 'compo', 'plus': ['净利润', '综合收益其他'], 'minus': []}}

>>> #查询证券利润表维度时间序列数据
>>> modelset.incdata('综合收益总额')
2010-03-31    0.079037
2010-06-30    0.143051
2010-09-30    0.120054
2010-12-31    0.122046
......
2017-03-31    0.346417
2017-06-30    0.472506
2017-09-30    0.275907
2017-12-31    0.275753
dtype: float64

>>> #维度模型样本内拟合
>>> modelset.fit('综合收益总额')
2014-09-30    0.324855
2014-12-31    0.235730
2015-03-31    0.193377
2015-06-30    0.343809
......
2017-03-31    0.349819
2017-06-30    0.580132
2017-09-30    0.468280
2017-12-31    0.404263
Freq: Q-DEC, dtype: float64

>>> #维度模型样本外推演
>>> modelset.forecast('综合收益总额',t=20)
2018-03-31    0.477166
2018-06-30    0.770036
2018-09-30    0.615717
2018-12-31    0.520084
......
2022-03-31    0.821030
2022-06-30    1.305306
2022-09-30    1.023223
2022-12-31    0.844383
Freq: Q-DEC, dtype: float64

>>> #现值计算
>>> modelset.pv('综合收益总额',r=0.1)
68.121251

>>> #交易日价格内部收益率计算
>>> modelset.irr('综合收益总额',price=68.121251,tradedate='2018-03-31')
0.1

>>> #构建测试交易日价格序列
>>> import pandas as pd
>>> import numpy as np
>>> priceseries=pd.Series([500,1000,5000,10000,50000],index=pd.date_range(start='2018-01-01',periods=5))
>>> #交易日价格序列内部收益率计算
>>> modelset.batchirr('综合收益总额',priceseries=priceseries)
                IRRs errors
2018-01-01  0.031065   None
2018-01-02  0.021390   None
2018-01-03  0.009234   None
2018-01-04  0.006475   None
2018-01-05       NaN  under

>>> #交易日价格兑现率计算
>>> modelset.achieverate('综合收益总额',price=68.121251,tradedate='2018-03-31',r=0.1)
1.0

>>> #交易日价格序列兑现率计算
>>> modelset.batchachieverate('综合收益总额',priceseries=priceseries,r=0.1)
2018-01-01    0.133070
2018-01-02    0.066553
2018-01-03    0.013314
2018-01-04    0.006659
2018-01-05    0.001332
Freq: D, dtype: float64

>>> #实际交易日价格内部收益率计算
>>> modelset.irr_trade('综合收益总额','close','2018-01-05')
0.235002

>>> #实际交易时期价格内部收益率计算
>>> modelset.batchirr_trade('综合收益总额',pricemode='close',startdate='2018-01-05',enddate='2018-01-10')
                IRRs errors
2018-01-05  0.235002   None
2018-01-08  0.227765   None
2018-01-09  0.224599   None
2018-01-10  0.220443   None

>>> #实际交易日价格兑现率计算
>>> modelset.achieverate_trade('综合收益总额',pricemode='close',tradedate='2017-03-16',r=0.1)
5.519836

>>> #实际交易日价格序列兑现率计算
>>> modelset.batchachieverate_trade('综合收益总额',pricemode='close',startdate='2017-03-10',enddate='2017-03-18',r=0.1)
2017-03-10    5.679002
2017-03-13    5.550267
2017-03-14    5.561767
2017-03-15    5.479098
2017-03-16    5.519836
2017-03-17    5.586371

>>> #使用TTM年度数据建模
>>> modelset=vq.StockModelSets.ma_strict_latest_status_modelset(stockcode='600690',structure=structure,reportdate='2017-12-31',mawindow=4,minstep=8,conservative=True,ttm=True)
>>> modelset.modelset()
{'营业收入': {'mode': 'auto', 'model': {'method': 'l1', 'step': 13, 'params': {'l1_diff': 1.60169, 'value_latest': 25.891691}, 'status': {'l1': '+', 'l2': 'n'}}}, '营业其他损益': {'mode': 'auto', 'model': {'method': 'l1', 'step': 13, 'params': {'l1_diff': 0.031299, 'value_latest': 0.491178}, 'status': {'l1': '+', 'l2': 'n'}}}, '营业外收入': {'mode': 'auto', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.06532}, 'status': {'l1': 'n'}}}, '营业外支出': {'mode': 'auto', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.013589}, 'status': {'l1': 'n'}}}, '利润总额其他': {'mode': 'auto', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.0}, 'status': {'l1': 'n'}}}, '综合收益其他': {'mode': 'auto', 'model': {'method': 'random', 'step': 7, 'params': {'value_mean': -0.00341}, 'status': {'l1': 'n'}}}, '营业成本': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.760509}, 'status': {'l1': 'n'}}}, '营业税金及附加': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.003877}, 'status': {'l1': 'n'}}}, '销售费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.128405}, 'status': {'l1': 'n'}}}, '管理费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.057998}, 'status': {'l1': 'n'}}}, '财务费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.000952}, 'status': {'l1': 'n'}}}, '营业其他费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 11, 'params': {'value_mean': 0.002096}, 'status': {'l1': 'n'}}}, '营业总收入': {'mode': 'compo', 'plus': ['营业收入'], 'minus': []}, '营业总成本': {'mode': 'compo', 'plus': ['营业成本', '营业税金及附加', '销售费用', '管理费用', '财务费用', '营业其他费用'], 'minus': []}, '营业利润': {'mode': 'compo', 'plus': ['营业总收入', '营业其他损益'], 'minus': ['营业总成本']}, '利润总额': {'mode': 'compo', 'plus': ['营业利润', '营业外收入', '利润总额其他'], 'minus': ['营业外支出']}, '税后利润': {'mode': 'quota', 'quota': 0.75, 'base': '利润总额'}, '净利润': {'mode': 'compo', 'plus': ['税后利润'], 'minus': []}, '综合收益总额': {'mode': 'compo', 'plus': ['净利润', '综合收益其他'], 'minus': []}}
>>> #设置参数ttm=True时，seasonal参数无意义
                            

>>> #构建模型集
>>> modelset=vq.StockFrameSets.ma_strict_latest_status_frameset(stockcode='600690',reportdate='2017-12-31',mawindow=4,minstep=8,seasonal=True,conservative=True)
>>> #获取建模结果
>>> modelset.modelset()
{'营业收入': {'mode': 'auto', 'model': {'method': 'l1', 'step': 14, 'params': {'l1_diff': 0.379913, 'value_latest': 7.85605}, 'status': {'l1': '+', 'l2': 'n'}, 'seasonal': {'12': 0.836401, '3': 0.963235, '6': 1.190104, '9': 0.977119}, 'qorder': ['3', '6', '9', '12']}}, '营业外收入': {'mode': 'auto', 'model': {'method': 'random', 'step': 32, 'params': {'value_mean': 0.024357}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.145846, '3': 0.769503, '6': 0.789491, '9': 0.952866}, 'qorder': ['3', '6', '9', '12']}}, '营业外支出': {'mode': 'auto', 'model': {'method': 'random', 'step': 58, 'params': {'value_mean': 0.001818}, 'status': {'l1': 'n'}, 'seasonal': {'12': 0.848634, '3': 0.602179, '6': 1.029571, '9': 1.205745}, 'qorder': ['3', '6', '9', '12']}}, '综合收益其他': {'mode': 'auto', 'model': {'method': 'random', 'step': 32, 'params': {'value_mean': -0.001163}, 'status': {'l1': 'n'}, 'seasonal': {'12': -0.094787, '3': 1.698016, '6': 1.446444, '9': 2.681057}, 'qorder': ['3', '6', '9', '12']}}, '营业成本': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 14, 'params': {'value_mean': 0.699182}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.002891, '3': 0.994847, '6': 0.988821, '9': 1.012805}, 'qorder': ['3', '6', '9', '12']}}, '营业税金及附加': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 58, 'params': {'value_mean': 0.003632}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.035196, '3': 0.840581, '6': 0.928841, '9': 1.04492}, 'qorder': ['3', '6', '9', '12']}}, '销售费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 45, 'params': {'value_mean': 0.137667}, 'status': {'l1': 'n'}, 'seasonal': {'12': 0.951441, '3': 1.088444, '6': 1.018624, '9': 0.943284}, 'qorder': ['3', '6', '9', '12']}}, '管理费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 23, 'params': {'value_mean': 0.068995}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.087347, '3': 1.017571, '6': 0.944272, '9': 0.942817}, 'qorder': ['3', '6', '9', '12']}}, '财务费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 58, 'params': {'value_mean': 0.000483}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.236724, '3': 1.338647, '6': 1.909793, '9': 0.460659}, 'qorder': ['3', '6', '9', '12']}}, '营业其他费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 48, 'params': {'value_mean': 0.001753}, 'status': {'l1': 'n'}, 'seasonal': {'12': 1.93244, '3': 1.349546, '6': 0.171696, '9': 0.833086}, 'qorder': ['3', '6', '9', '12']}}, '营业总收入': {'mode': 'compo', 'plus': ['营业收入'], 'minus': []}, '营业总成本': {'mode': 'compo', 'plus': ['营业成本', '营业税金及附加', '销售费用', '管理费用', '财务费用', '营业其他费用'], 'minus': []}, '营业利润': {'mode': 'compo', 'plus': ['营业总收入'], 'minus': ['营业总成本']}, '利润总额': {'mode': 'compo', 'plus': ['营业利润', '营业外收入'], 'minus': ['营业外支出']}, '税后利润': {'mode': 'quota', 'quota': 0.75, 'base': '利润总额'}, '净利润': {'mode': 'compo', 'plus': ['税后利润'], 'minus': []}, '综合收益总额': {'mode': 'compo', 'plus': ['净利润', '综合收益其他'], 'minus': []}}

>>> #查询证券利润表维度时间序列数据
>>> modelset.incdata('综合收益总额')
2010-03-31    0.079037
2010-06-30    0.143051
2010-09-30    0.120054
2010-12-31    0.122046
......
2017-03-31    0.346417
2017-06-30    0.472506
2017-09-30    0.275907
2017-12-31    0.275753
dtype: float64

>>> #维度模型样本内拟合
>>> modelset.fit('综合收益总额')
2014-09-30    0.324855
2014-12-31    0.235730
2015-03-31    0.193377
2015-06-30    0.343809
......
2017-03-31    0.349819
2017-06-30    0.580132
2017-09-30    0.468280
2017-12-31    0.404263
Freq: Q-DEC, dtype: float64

>>> #维度模型样本外推演
>>> modelset.forecast('综合收益总额',t=20)
2018-03-31    0.477166
2018-06-30    0.770036
2018-09-30    0.615717
2018-12-31    0.520084
......
2022-03-31    0.821030
2022-06-30    1.305306
2022-09-30    1.023223
2022-12-31    0.844383
Freq: Q-DEC, dtype: float64

>>> #现值计算
>>> modelset.pv('综合收益总额',r=0.1)
68.121251

>>> #交易日价格内部收益率计算
>>> modelset.irr('综合收益总额',price=68.121251,tradedate='2018-03-31')
0.1

>>> #构建测试交易日价格序列
>>> import pandas as pd
>>> import numpy as np
>>> priceseries=pd.Series([500,1000,5000,10000,50000],index=pd.date_range(start='2018-01-01',periods=5))
>>> #交易日价格序列内部收益率计算
>>> modelset.batchirr('综合收益总额',priceseries=priceseries)
                IRRs errors
2018-01-01  0.031065   None
2018-01-02  0.021390   None
2018-01-03  0.009234   None
2018-01-04  0.006475   None
2018-01-05       NaN  under

>>> #交易日价格兑现率计算
>>> modelset.achieverate('综合收益总额',price=68.121251,tradedate='2018-03-31',r=0.1)
1.0

>>> #交易日价格序列兑现率计算
>>> modelset.batchachieverate('综合收益总额',priceseries=priceseries,r=0.1)
2018-01-01    0.133070
2018-01-02    0.066553
2018-01-03    0.013314
2018-01-04    0.006659
2018-01-05    0.001332
Freq: D, dtype: float64

>>> #实际交易日价格内部收益率计算
>>> modelset.irr_trade('综合收益总额','close','2018-01-05')
0.235002

>>> #实际交易时期价格内部收益率计算
>>> modelset.batchirr_trade('综合收益总额',pricemode='close',startdate='2018-01-05',enddate='2018-01-10')
                IRRs errors
2018-01-05  0.235002   None
2018-01-08  0.227765   None
2018-01-09  0.224599   None
2018-01-10  0.220443   None

>>> #实际交易日价格兑现率计算
>>> modelset.achieverate_trade('综合收益总额',pricemode='close',tradedate='2017-03-16',r=0.1)
5.519836

>>> #实际交易日价格序列兑现率计算
>>> modelset.batchachieverate_trade('综合收益总额',pricemode='close',startdate='2017-03-10',enddate='2017-03-18',r=0.1)
2017-03-10    5.679002
2017-03-13    5.550267
2017-03-14    5.561767
2017-03-15    5.479098
2017-03-16    5.519836
2017-03-17    5.586371

>>> #使用TTM年度数据建模
>>> modelset=vq.StockFrameSets.ma_strict_latest_status_frameset(stockcode='600690',reportdate='2017-12-31',mawindow=4,minstep=8,conservative=True,ttm=True)
>>> modelset.modelset()
{'营业收入': {'mode': 'auto', 'model': {'method': 'l1', 'step': 13, 'params': {'l1_diff': 1.60169, 'value_latest': 25.891691}, 'status': {'l1': '+', 'l2': 'n'}}}, '营业其他损益': {'mode': 'auto', 'model': {'method': 'l1', 'step': 13, 'params': {'l1_diff': 0.031299, 'value_latest': 0.491178}, 'status': {'l1': '+', 'l2': 'n'}}}, '营业外收入': {'mode': 'auto', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.06532}, 'status': {'l1': 'n'}}}, '营业外支出': {'mode': 'auto', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.013589}, 'status': {'l1': 'n'}}}, '利润总额其他': {'mode': 'auto', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.0}, 'status': {'l1': 'n'}}}, '综合收益其他': {'mode': 'auto', 'model': {'method': 'random', 'step': 7, 'params': {'value_mean': -0.00341}, 'status': {'l1': 'n'}}}, '营业成本': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.760509}, 'status': {'l1': 'n'}}}, '营业税金及附加': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.003877}, 'status': {'l1': 'n'}}}, '销售费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.128405}, 'status': {'l1': 'n'}}}, '管理费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.057998}, 'status': {'l1': 'n'}}}, '财务费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 13, 'params': {'value_mean': 0.000952}, 'status': {'l1': 'n'}}}, '营业其他费用': {'mode': 'derive', 'base': '营业收入', 'model': {'method': 'random', 'step': 11, 'params': {'value_mean': 0.002096}, 'status': {'l1': 'n'}}}, '营业总收入': {'mode': 'compo', 'plus': ['营业收入'], 'minus': []}, '营业总成本': {'mode': 'compo', 'plus': ['营业成本', '营业税金及附加', '销售费用', '管理费用', '财务费用', '营业其他费用'], 'minus': []}, '营业利润': {'mode': 'compo', 'plus': ['营业总收入', '营业其他损益'], 'minus': ['营业总成本']}, '利润总额': {'mode': 'compo', 'plus': ['营业利润', '营业外收入', '利润总额其他'], 'minus': ['营业外支出']}, '税后利润': {'mode': 'quota', 'quota': 0.75, 'base': '利润总额'}, '净利润': {'mode': 'compo', 'plus': ['税后利润'], 'minus': []}, '综合收益总额': {'mode': 'compo', 'plus': ['净利润', '综合收益其他'], 'minus': []}}
>>> #设置参数ttm=True时，seasonal参数无意义