股市风起云涌，我用Python分析周期之道-经管之家官网！

1. def Return_Dates(forecast,stock_data,cycle,cycle_name = 'self_define_cycle',time_name = 'ds'):
   
2. # find out the highest and lowest dates in the first cycle
   
3. # We cannot simply search for all highest and lowest point since there is slightly difference for high and low values in different cycles
   
4. high = forecast.iloc[:cycle,]
   
5. high = high[high[cycle_name]==max(high[cycle_name])][time_name]
   
6. high = datetime.strptime(str(high.values[0])[:10],"%Y-%m-%d")
   
7. low = forecast.iloc[:cycle,]
   
8. low = low[low[cycle_name]==min(low[cycle_name])][time_name]
   
9. low = datetime.strptime(str(low.values[0])[:10],"%Y-%m-%d")
   
10. end_dt = datetime.strptime(stock_data.index[-1],"%Y-%m-%d")
    
11. find_list = stock_data.index.map(lambda x:datetime.strptime(x,"%Y-%m-%d"))
    
12. # Finding selling and buying dates with loop
    
13. sell_dt = []
    
14. sell_dt.append(high)
    
15. # Looking for new cycle until it goes beyond the last date in stock_data
    
16. while high<end_dt:
    
17. high = high+timedelta(days=cycle)
    
18. dif = (find_list-high).days
    
19. high = find_list[abs(dif)==min(abs(dif))][0] # In order to avoid the non-trading dates
    
20. sell_dt.append(high)
    
21. buy_dt = []
    
22. buy_dt.append(low)
    
23. # Looking for new cycle until it goes beyond the last date in stock_data
    
24. while low<end_dt:
    
25. low = low+timedelta(days=cycle)
    
26. dif = (find_list-low).days
    
27. low = find_list[abs(dif)==min(abs(dif))][0] # In order to avoid the non-trading dates
    
28. buy_dt.append(low)
    
29. if buy_dt[0] > sell_dt[0]:
    
30. sell_dt = sell_dt[1:]
    
31. buy_dt = buy_dt[:-1]
    
32. sell_dt = sell_dt[:-1]
    
33. return [buy_dt,sell_dt]

在2015/10/1和2018/10/1期间，我们买卖Costco四次。3年内的回报率为23.2％。 可能不是很吸引人，但至少它是比较乐观的回报。

https://user-gold-cdn.xitu.io/2018/10/16/1667accbca7fb022?imageView2/0/w/1280/h/960/format/webp/ignore-error/1

更多股票的应用,当然，这种方法可以应用于尽可能多的股票。 我们列出了Costco，Apple，Microsoft，Home Depot和Nike的平均购买价格，平均销售价格，周期长度，样本均方误差，购买数量，销售数量和每个周期内的预计回报。

1. Analysis_ticks = ['COST','AAPL','MSFT','HD','NKE']
   
2. start_date = '2015-10-01'
   
3. end_date = '2018-10-01'
   
4. opt_cycle = []
   
5. prot_return = []
   
6. MSE = []
   
7. buy_times = []
   
8. sell_times = []
   
9. avg_buy_price = []
   
10. avg_sell_price = []
    
11. # Loop over each stock
    
12. for ticker in Analysis_ticks:
    
13. stock_data = data.DataReader(ticker, 'iex', start_date, end_date)
    
14. testing_box = range(50,301)
    
15. return_box = []
    
16. mse_box = []
    
17. for cc in testing_box:
    
18. f = cycle_analysis(stock_data['close'],'2018-04-01',cc)
    
19. return_box.append(f[0])
    
20. mse_box.append(f[1])
    
21. report = pd.DataFrame({'cycle':testing_box,'return':return_box,'mse':mse_box})
    
22. possible_choice = report[report['return'] >10]
    
23. # If we cannot find a cycle with return greater than 10, give 0
    
24. if possible_choice.shape[0]>0:
    
25. c = possible_choice[possible_choice['mse']==min(possible_choice['mse'])]['cycle'].values[0]
    
26. rp = possible_choice[possible_choice['mse']==min(possible_choice['mse'])]['return'].values[0]
    
27. mse = possible_choice[possible_choice['mse']==min(possible_choice['mse'])]['mse'].values[0]
    
28. df = stock_data[:'2018-04-01'].iloc[:-1,]['close'].reset_index()
    
29. df.columns = ['ds','y']
    
30. predict_period = len(pd.date_range('2018-04-01','2018-10-01'))
    
31. m = Prophet(weekly_seasonality=False,yearly_seasonality=False,daily_seasonality=False)
    
32. m.add_seasonality('self_define_cycle',period=c,fourier_order=8,mode='additive')
    
33. m.fit(df)
    
34. future = m.make_future_dataframe(periods=predict_period)
    
35. forecast = m.predict(future)
    
36. dt_list = Return_Dates(forecast,stock_data,c)
    
37. buy_price = stock_data.loc[map(lambda x: x.strftime("%Y-%m-%d"),dt_list[0])]['close']
    
38. sell_price = stock_data.loc[map(lambda x: x.strftime("%Y-%m-%d"),dt_list[1])]['close']
    
39. bt = buy_price.shape[0]
    
40. st = sell_price.shape[0]
    
41. bp = np.mean(buy_price)
    
42. sp = np.mean(sell_price)
    
43. else:
    
44. c = 0
    
45. rp = 0
    
46. mse = 0
    
47. bt = 0
    
48. st = 0
    
49. bp = 0
    
50. sp = 0
    
51. opt_cycle.append(c)
    
52. prot_return.append(rp)
    
53. MSE.append(mse)
    
54. buy_times.append(bt)
    
55. sell_times.append(st)
    
56. avg_buy_price.append(bp)
    
57. avg_sell_price.append(sp)
    
58. print "{} Finished".format(ticker)

复制代码

对于微软和耐克，我们找不到符合我们要求每个周期超过10元回报的周期。 对于Costco，Apple和Home Depot，我们可以找到大约250天的周期，并做出良好的预测和良好的回报。

1. stock_report = pd.DataFrame({'Stock':Analysis_ticks,'Cycle':opt_cycle,'Projected_Return_per_Cycle':prot_return,
   
2. 'MSE':MSE,'Num_of_Buy':buy_times,'Num_of_Sell':sell_times,
   
3. 'Average_Buy_Price':avg_buy_price,'Average_Sell_Price':avg_sell_price})
   
4. stock_report

复制代码

https://user-gold-cdn.xitu.io/2018/10/16/1667acd6b0cad10f?w=1080&h=155&f=jpeg&s=35581

总结

借助Python和fbprophet包，我们可以更好地了解股市。 根据我们发现的周期，我们可以在3年内获得大约23％的回报。 也许这种投资策略无法满足所有人的需求，但你始终可以根据自己的知识和经验设定自己的方法。 强大的fbprophet软件包可以让你对股票市场的分析更加深入和轻松。

延伸阅读：

经典的深度学习教程及Python代码实现

「经管之家」APP：经管人学习、答疑、交友，就上经管之家！
免流量费下载资料----在经管之家app可以下载论坛上的所有资源，并且不额外收取下载高峰期的论坛币。
涵盖所有经管领域的优秀内容----覆盖经济、管理、金融投资、计量统计、数据分析、国贸、财会等专业的学习宝库，各类资料应有尽有。
来自五湖四海的经管达人----已经有上千万的经管人来到这里，你可以找到任何学科方向、有共同话题的朋友。
经管之家（原人大经济论坛），跨越高校的围墙，带你走进经管知识的新世界。
扫描下方二维码下载并注册APP