Question

尝试将列添加到pandas数据帧中，该数据帧是使用pyvollib（from py_vollib.black.implied_volatility import implied_volatility）中的implied_volatility（）函数计算的。所有输入参数均来自以下df：

In [101]: odf.head()
Out[101]: 
       Expiry      K Type     close         F  texp         r
7  2018-01-26  120.5    p  0.015625  122.9375    10  0.012604
10 2018-01-26  121.0    p  0.015625  122.9375    10  0.012604
11 2018-01-26  122.0    p  0.062500  122.9375    10  0.012604
17 2018-01-26  121.5    p  0.015625  122.9375    10  0.012604
19 2018-01-26  122.5    p  0.140625  122.9375    10  0.012604

我正在使用的功能：

odf['iv'] = [implied_volatility(discounted_option_price = odf.close,F=odf.F,K=odf.K,
                 r=odf.r,t=odf.texp/365,flag=odf.Type)
                 for rows in odf]

问题似乎是odf.Type列，它是一系列字符串。

我在ipython中获取回溯：

In [102]: odf['iv'] = [implied_volatility(discounted_option_price = odf.close,F=
     ...: odf.F,K=odf.K,
     ...:                  r=odf.r,t=odf.texp/365,flag=odf.Type)
     ...:                  for rows in odf]
     ...: 
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-102-133149ca6fb0> in <module>()
      1 odf['iv'] = [implied_volatility(discounted_option_price = odf.close,F=odf.F,K=odf.K,
      2                  r=odf.r,t=odf.texp/365,flag=odf.Type)
----> 3                  for rows in odf]

<ipython-input-102-133149ca6fb0> in <listcomp>(.0)
      1 odf['iv'] = [implied_volatility(discounted_option_price = odf.close,F=odf.F,K=odf.K,
      2                  r=odf.r,t=odf.texp/365,flag=odf.Type)
----> 3                  for rows in odf]

~/miniconda3/lib/python3.6/site-packages/py_vollib/black/implied_volatility.py in implied_volatility_of_discounted_option_price(discounted_option_price, F, K, r, t, flag)
     97         K,
     98         t,
---> 99         binary_flag[flag]
    100     )
    101     if sigma_calc == FLOAT_MAX:

~/miniconda3/lib/python3.6/site-packages/pandas/core/generic.py in __hash__(self)
   1043     def __hash__(self):
   1044         raise TypeError('{0!r} objects are mutable, thus they cannot be'
-> 1045                         ' hashed'.format(self.__class__.__name__))
   1046 
   1047     def __iter__(self):

TypeError: 'Series' objects are mutable, thus they cannot be hashed

任何指针如何让这个运行将非常感激。

Answer 1

我相信你需要apply：

odf['iv'] = odf.apply(lambda x: implied_volatility(discounted_option_price = x.close,
                                                   F=x.F,
                                                   K=x.K,
                                                   r=x.r,
                                                   t=x.texp/365,
                                                   flag=x.Type), axis=1)

或者：

odf['iv'] = odf.apply(lambda x: implied_volatility(x.close,
                                                   x.F,
                                                   x.K,
                                                   x.r,
                                                   x.texp/365,
                                                   x.Type), axis=1)

Answer 2

odf['iv'] = [implied_volatility(discounted_option_price = odf.loc[i,'close'],F=odf.loc[i,'F'],K=odf.loc[i,'k'],
                 r=odf.loc[i,'r'],t=odf.loc[i,'texp']/365,flag=odf.loc[i,'Type'])
                 for i in odf.index]

使用apply可能有一些更好的方法，但我不知道它如何用于命名参数。

需要使用包中的函数向pd.df添加列 - TypeError：'Series'对象是可变的

2 个答案: