# list_status L:上市

all_stocks = get_tushare_api().stock_basic(list_status='L', fields='ts_code,name,industry')

# 收益率列表

pf_list = []

cnt = 0

for index, row in all_stocks.iterrows():

try:

day_two = 2

# 5日

day_five = 5

# 10日

day_ten = 10

# 50日

day_fifty = 50

# 120日

day_one_hundred_twenty = 120

# 250日

day_year = 250

# code = row['ts_code']

# df = get_tushare_api().daily( ts_code=row['ts_code'], start_date=start_date, end_date=end_date)

code = row['ts_code'][:-3]

df = ts.get_k_data(code, start=start_date, end=end_date[0], ktype="D", autype="qfq") # 读日K行情数据

# 上市时间大于4年的不看

if len(df) > 1000:

continue

df.index = df.pop("date")

if (len(df) < day_ten):

continue

if (len(df) < day_fifty or len(df) < day_one_hundred_twenty or len(df) < day_year):

day_fifty = day_one_hundred_twenty = day_year = len(df)

# 最近50日的收益率

profit = int((df["close"][-1] / df["close"][-day_fifty] - 1.0) * 100)

# 2日线

df["two"] = df['close'].rolling(day_two).mean()

df["two"] = round(df["two"], 2)

# 5日线

df["five"] = df['close'].rolling(day_five).mean()

df["five"] = round(df["five"], 2)

# 10日线

df["ten"] = df['close'].rolling(day_ten).mean()

df["ten"] = round(df["ten"], 2)

# 50 日线

df['fifty'] = df['close'].rolling(day_fifty).mean()

df["fifty"] = round(df["fifty"], 2)

# 50日最高

df['fifty-max'] = df['close'].rolling(day_fifty).max()

df['fifty-max'] = round(df['fifty-max'], 2)

# 50 -1日最低

df['forty-nine-min'] = round(df['close'].rolling(day_fifty).min(), 2)

# 120日最高价

df["day_one_hundred_twenty"] = round(df['close'].rolling(day_one_hundred_twenty).max(), 2)

# 年线 250日

df["day_year"] = round(df['close'].rolling(day_year).mean(), 2)

# 是否满足反转条件 日线收盘价站上年线； 一月内曾创50日新高；收盘价站上年线的天数大于3，小于30

# 50日内振幅小于42% ((最高价-最低价)/最低价) 且 昨天为1个半月内最高价；

df["buy"] = (df["close"] > df["day_year"]) & \

(df["close"].rolling(30).max() >= df['fifty']) & \

(df["close"].rolling(3).mean() > df["day_year"]) & \

((df["fifty-max"] - df['forty-nine-min']) / df['forty-nine-min'] <= 0.45) & \

(round(df["close"], 2) == df['fifty-max'])

signals = df[df["buy"]]

if len(signals):

rev = True

date = min(list(signals.index))

else:

rev = False

date = ""

pf_list.append(dict(code=code, pf=profit, name=row['name'], industry=row['industry'], revs=rev, date=date))

cnt += 1

except Exception as e:

print("%s 出错,错误详情:[%s]" % (code, e))

continue

"""

陶博士月线翻转策略

月线反转3.0版本的技术指标公式的几个条件是：

(1)日线收盘价站上年线；

(2)一月内曾创50日新高；

(3)50日的RPS大于85；

(4)收盘价站上年线的天数大于3，小于30；

(5)最高价距离120日内的最高价不到10%；

"""

api = get_tushare_api()

# 取收益最高的15%，计算收益排名的均值 牛市选择50%

cmp_percent = 0.50

end_date_format = end_date[0]

pf_list = []

if (os.path.isfile("strategy/" + end_date_format)):

with open("strategy/" + end_date_format, "r", encoding="utf-8") as dt:

for line in dt.readlines():

line = line.strip('\n')

pf_list.append(eval(line))

pf_list.sort(key=lambda x: x["pf"], reverse=True) # 先按时间内的收益率排名（从高往低）

else:

pf_list = get_list(start_date, end_date)

pf_list.sort(key=lambda x: x["pf"], reverse=True) # 先按时间内的收益率排名（从高往低）

with open("strategy/" + end_date_format, 'a', encoding='utf-8') as f:

for item in pf_list:

f.write("%s\n" % item)

for idx, doc in enumerate(pf_list):

doc["rank"] = idx # 记录下每只个股在时间内的收益率排名

num_cands = int(len(pf_list) * cmp_percent) # 用于统计的最高（和最低）收益的股票个数

top_cands = pf_list[:num_cands] # 时间内最高收益的股票信息列表

for t_idx, t_doc in enumerate(top_cands):

# 是否月线反转

if t_doc['revs']:

# 反转日期

appear_date = t_doc['date'].replace("-", "")

# 判断日期是否为1个月内

timeDay = datetime.now() - timedelta(days=30)

str_date = timeDay.strftime("%Y%m%d")

# 反转日期小于1个月前日期

if (datetime.strptime(appear_date, '%Y%m%d') < datetime.strptime(str_date, '%Y%m%d')):

continue

name = t_doc['name']

# number = get_hk_number(str_date,name)

# if (number >= 100):

timeDay = datetime.now() - timedelta(days=date)

if (t_doc['date'] >= timeDay.strftime("%Y-%m-%d")):

# 获取当日涨幅

day_format = t_doc['date'].replace("-", "")

day_detail = api.daily(ts_code=get_code(t_doc['code']), start_date=day_format,

end_date=day_format)

print(t_doc['code'], t_doc['name'], t_doc['rank'], t_doc['date'], t_doc['industry'],

with open("date/" + date, 'r', encoding='utf-8') as d:

data_str = d.read()

j = json.loads(data_str)

df = pd.DataFrame(j, columns=['code', 'name', 'number', 'ratio'])

curr = (df.loc[df['name'] == name])

if len(curr) == 0:

return

'''

获取业绩 如果有业绩则 大于等于min最小增长率为筛选条件

:param date: 日期

:param min: 最小增长值

:return:

'''

pro = get_tushare_api()

forecast = pro.forecast(period=date)

for index, name in forecast.iterrows():

if (name['p_change_min'] > min):

print("code: %s " % name['ts_code'], name['end_date'], name['type'], "最小值: %s " % name['p_change_min'],

name['last_parent_net'],

name['summary'], name['change_reason'])

else:

print(name['ts_code'], name['end_date'], name['type'], name['p_change_min'], name['last_parent_net'],

if code[:2] == '60' or code[:2] == '68':

return code + '.SH'

else: