Python量化教程 常用函数

# -*- coding: utf-8 -*-

# @Author: fangbei

# @Date:  2017-08-26

# @Original:

price_str = '30.14, 29.58, 26.36, 32.56, 32.82'

price_str = price_str.replace(' ', '')  #删除空格

price_array = price_str.split(',')      #转成数组

date_array = []

date_base = 20170118

'''

# for 循环

for _ in range(0, len(price_array)):

    date_array.append(str(date_base))

    date_base += 1

'''

#推导式comprehensions（又称解析式），是Python的一种独有特性。推导式是可以从一个数据序列构建另一个新的数据序列的结构体。

#列表推导式

date_array = [str(date_base + ind) for ind, _ in enumerate(price_array)]

print(date_array)

# ['20170118', '20170119', '20170120', '20170121', '20170122']

# zip函数

stock_tuple_list = [(date, price) for date, price in zip(date_array, price_array)]

print(stock_tuple_list)

# [('20170118', '30.14'), ('20170119', '29.58'), ('20170120', '26.36'), ('20170121', '32.56'), ('20170122', '32.82')]

#字典推导式

stock_dict = {date: price for date, price in zip(date_array, price_array)}

print(stock_dict)

# {'20170118': '30.14', '20170119': '29.58', '20170120': '26.36', '20170121': '32.56', '20170122': '32.82'}

# 可命名元组 namedtuple

from collections import namedtuple

stock_nametuple = namedtuple('stock', ('date', 'price'))

stock_nametuple_list = [stock_nametuple(date, price) for date, price in zip(date_array, price_array)]

print(stock_nametuple_list)

# [stock(date='20170118', price='30.14'), stock(date='20170119', price='29.58'), stock(date='20170120', price='26.36'), stock(date='20170121', price='32.56'), stock(date='20170122', price='32.82')]

# 有序字典 OrderedDict

from collections import OrderedDict

stock_dict = OrderedDict((date, price) for date, price in zip(date_array, price_array))

print(stock_dict.keys())

# odict_keys(['20170118', '20170119', '20170120', '20170121', '20170122'])

#最小收盘价

print(min(zip(stock_dict.values(), stock_dict.keys())))

# ('26.36', '20170120')

#lambad函数

func = lambda x:x+1

#以上lambda等同于以下函数

def func(x):

    return(x+1)

#找出收盘价中第二大的价格

find_second_max_lambda = lambda dict_array : sorted(zip(dict_array.values(), dict_array.keys()))[-2]

print(find_second_max_lambda(stock_dict))

# ('32.56', '20170121')

#高阶函数

#将相邻的收盘价格组成tuple后装入list

price_float_array = [float(price_str) for price_str in stock_dict.values()]

pp_array = [(price1, price2) for price1, price2 in zip(price_float_array[:-1], price_float_array[1:])]

print(pp_array)

# [(30.14, 29.58), (29.58, 26.36), (26.36, 32.56), (32.56, 32.82)]

from functools import reduce

#外层使用map函数针对pp_array()的每一个元素执行操作，内层使用reduce()函数即两个相邻的价格， 求出涨跌幅度，返回外层结果list

change_array = list(map(lambda pp:reduce(lambda a,b: round((b-a) / a, 3),pp), pp_array))

# print(type(change_array))

change_array.insert(0,0)

print(change_array)

# [0, -0.019, -0.109, 0.235, 0.008]

#将涨跌幅数据加入OrderedDict，配合使用namedtuple重新构建数据结构stock_dict

stock_nametuple = namedtuple('stock', ('date', 'price', 'change'))

stock_dict = OrderedDict((date, stock_nametuple(date, price, change))

                        for date, price, change in

                        zip(date_array, price_array, change_array))

print(stock_dict)

# OrderedDict([('20170118', stock(date='20170118', price='30.14', change=0)), ('20170119', stock(date='20170119', price='29.58', change=-0.019)), ('20170120', stock(date='20170120', price='26.36', change=-0.109)), ('20170121', stock(date='20170121', price='32.56', change=0.235)), ('20170122', stock(date='20170122', price='32.82', change=0.008))])

#用filter()进行筛选，选出上涨的交易日

up_days = list(filter(lambda day: day.change > 0, stock_dict.values()))

print(up_days)

# [stock(date='20170121', price='32.56', change=0.235), stock(date='20170122', price='32.82', change=0.008)]

#定义函数计算涨跌日或涨跌值

def filter_stock(stock_array_dict, want_up=True, want_calc_sum=False):

    if not isinstance(stock_array_dict, OrderedDict):

        raise TypeError('stock_array_dict must be OrderedDict')

    filter_func = (lambda day: day.change > 0) if want_up else (lambda day: day.change < 0)

    want_days = list(filter(filter_func, stock_array_dict.values()))

    if not want_calc_sum:

        return want_days

    change_sum = 0.0

    for day in want_days:

        change_sum += day.change

    return change_sum

#偏函数 partial

from functools import partial

filter_stock_up_days    = partial(filter_stock, want_up=True,  want_calc_sum=False)

filter_stock_down_days  = partial(filter_stock, want_up=False, want_calc_sum=False)

filter_stock_up_sums    = partial(filter_stock, want_up=True,  want_calc_sum=True)

filter_stock_down_sums  = partial(filter_stock, want_up=False, want_calc_sum=True)

print('所有上涨的交易日：{}'.format(list(filter_stock_up_days(stock_dict))))

print('所有下跌的交易日：{}'.format(list(filter_stock_down_days(stock_dict))))

print('所有上涨交易日的涨幅和：{}'.format(filter_stock_up_sums(stock_dict)))

print('所有下跌交易日的跌幅和：{}'.format(filter_stock_down_sums(stock_dict)))

# 所有上涨的交易日：[stock(date='20170121', price='32.56', change=0.235), stock(date='20170122', price='32.82', change=0.008)]

# 所有下跌的交易日：[stock(date='20170119', price='29.58', change=-0.019), stock(date='20170120', price='26.36', change=-0.109)]

# 所有上涨交易日的涨幅和：0.243

# 所有下跌交易日的跌幅和：-0.128

来源：博客园    作者：比特量化

------------------------------------------------------------------------------------------------

拓展阅读：如何使用Python实现你的股票量化交易模型