Market Competitiveness · Python 量化交易教程

# Market Competitiveness > 來源：https://uqer.io/community/share/54b5c2f1f9f06c276f651a17 來一個奇葩無厘頭的市場競爭策略 ## 策略思路某一行業的幾大龍頭股票，在穩定時期此消彼長 ## 策略實現 + 股票池：選擇一行業內的流動性比較好的龍頭股票；例如三家自助品牌汽車，長安、比亞迪和長城，以下按照三只股票情況討論 + 觀察某一天時，股票價格和該股票在過去幾天內平均值的關系 + 如果兩只股票下跌，則預測另一只股票上漲；如果兩只股票上漲，則預測另一只股票下跌 + 如果某天三只股票中的兩只較其平均值有較大幅度下跌，而另一只股票較其平均值比較穩定不變，則買入后面這只比較穩定的股票 + 如果某天三只股票中的兩只較其平均值有較大幅度上漲，而另一只股票較其平均值比較穩定不變，則賣出后面這只比較穩定的股票 ```py import quartz import quartz.backtest as qb import quartz.performance as qp from quartz.api import * import pandas as pd import numpy as np from datetime import datetime from matplotlib import pylab ``` ```py start = datetime(2012, 1, 1) end = datetime(2014, 12, 1) benchmark = 'HS300' universe = ['000625.XSHE', # 長安汽車 '002594.XSHE', # 比亞迪汽車 '601633.XSHG' # 長城汽車 ] capital_base = 1000000 refresh_rate = 5 window = 10 def initialize(account): account.amount = 100000 account.universe = universe add_history('hist', window) def handle_data(account): stk_0 = universe[0] stk_1 = universe[1] stk_2 = universe[2] prices_0 = account.hist[stk_0]['closePrice'] prices_1 = account.hist[stk_1]['closePrice'] prices_2 = account.hist[stk_2]['closePrice'] mu_0 = prices_0.mean() mu_1 = prices_1.mean() mu_2 = prices_2.mean() # 兩只下跌較大幅度，一只較穩定，買入較穩定這只股票 if prices_0[-1] > mu_0 and prices_1[-1] < 0.975 * mu_1 and prices_2[-1] < 0.975 * mu_2: order(stk_0, account.amount) if prices_1[-1] > mu_1 and prices_2[-1] < 0.975 * mu_2 and prices_0[-1] < 0.975 * mu_0: order(stk_1, account.amount) if prices_2[-1] > mu_2 and prices_0[-1] < 0.975 * mu_0 and prices_1[-1] < 0.975 * mu_1: order(stk_2, account.amount) # 兩只上漲較大幅度，一只較穩定，賣出較穩定這只股票 if prices_0[-1] < mu_0 and prices_1[-1] > 1.025 * mu_1 and prices_2[-1] > 1.025 * mu_2: order_to(stk_0, 0) if prices_1[-1] < mu_1 and prices_0[-1] > 1.025 * mu_0 and prices_2[-1] > 1.025 * mu_2: order_to(stk_1, 0) if prices_2[-1] < mu_2 and prices_0[-1] > 1.025 * mu_0 and prices_1[-1] > 1.025 * mu_1: order_to(stk_2, 0) ``` ![](https://box.kancloud.cn/2016-07-30_579cbdb33b9d8.jpg) ```py bt ``` | | tradeDate | cash | stock_position | portfolio_value | benchmark_return | blotter | | --- | --- | | 0 | 2012-01-18 | 1000000.00000 | {} | 1000000.00000 | 0.000000 | [] | | 1 | 2012-01-19 | 1000000.00000 | {} | 1000000.00000 | 0.019058 | [] | | 2 | 2012-01-20 | 1000000.00000 | {} | 1000000.00000 | 0.014478 | [] | | 3 | 2012-01-30 | 1000000.00000 | {} | 1000000.00000 | -0.017318 | [] | | 4 | 2012-01-31 | 1000000.00000 | {} | 1000000.00000 | 0.001439 | [] | | 5 | 2012-02-01 | 1000000.00000 | {} | 1000000.00000 | -0.014311 | [] | | 6 | 2012-02-02 | 1000000.00000 | {} | 1000000.00000 | 0.023567 | [] | | 7 | 2012-02-03 | 1000000.00000 | {} | 1000000.00000 | 0.007985 | [] | | 8 | 2012-02-06 | 1000000.00000 | {} | 1000000.00000 | -0.000705 | [] | | 9 | 2012-02-07 | 1000000.00000 | {} | 1000000.00000 | -0.018515 | [] | | 10 | 2012-02-08 | 1000000.00000 | {} | 1000000.00000 | 0.028594 | [] | | 11 | 2012-02-09 | 1000000.00000 | {} | 1000000.00000 | 0.000394 | [] | | 12 | 2012-02-10 | 1000000.00000 | {} | 1000000.00000 | 0.001737 | [] | | 13 | 2012-02-13 | 1000000.00000 | {} | 1000000.00000 | -0.000648 | [] | | 14 | 2012-02-14 | 1000000.00000 | {} | 1000000.00000 | -0.003900 | [] | | 15 | 2012-02-15 | 1000000.00000 | {} | 1000000.00000 | 0.010904 | [] | | 16 | 2012-02-16 | 1000000.00000 | {} | 1000000.00000 | -0.005308 | [] | | 17 | 2012-02-17 | 1000000.00000 | {} | 1000000.00000 | 0.000399 | [] | | 18 | 2012-02-20 | 1000000.00000 | {} | 1000000.00000 | 0.001427 | [] | | 19 | 2012-02-21 | 1000000.00000 | {} | 1000000.00000 | 0.008559 | [] | | 20 | 2012-02-22 | 1000000.00000 | {} | 1000000.00000 | 0.013668 | [] | | 21 | 2012-02-23 | 1000000.00000 | {} | 1000000.00000 | 0.003380 | [] | | 22 | 2012-02-24 | 1000000.00000 | {} | 1000000.00000 | 0.016023 | [] | | 23 | 2012-02-27 | 1000000.00000 | {} | 1000000.00000 | 0.003231 | [] | | 24 | 2012-02-28 | 1000000.00000 | {} | 1000000.00000 | 0.002217 | [] | | 25 | 2012-02-29 | 1000000.00000 | {} | 1000000.00000 | -0.010637 | [] | | 26 | 2012-03-01 | 1000000.00000 | {} | 1000000.00000 | -0.000303 | [] | | 27 | 2012-03-02 | 1000000.00000 | {} | 1000000.00000 | 0.017692 | [] | | 28 | 2012-03-05 | 1000000.00000 | {} | 1000000.00000 | -0.006432 | [] | | 29 | 2012-03-06 | 1000000.00000 | {} | 1000000.00000 | -0.015641 | [] | | ... | ... | ... | ... | ... | ... | ... | | 664 | 2014-10-21 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1913031.23401 | -0.008685 | [] | | 665 | 2014-10-22 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1933648.47401 | -0.006062 | [] | | 666 | 2014-10-23 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1953640.67401 | -0.009385 | [] | | 667 | 2014-10-24 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1823065.79401 | -0.002183 | [] | | 668 | 2014-10-27 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1859302.04401 | -0.009152 | [] | | 669 | 2014-10-28 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1863675.44401 | 0.020187 | [] | | 670 | 2014-10-29 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1871797.24401 | 0.014371 | [] | | 671 | 2014-10-30 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1883042.97401 | 0.007156 | [] | | 672 | 2014-10-31 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1913656.09401 | 0.015958 | [] | | 673 | 2014-11-03 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1902410.64401 | 0.001682 | [] | | 674 | 2014-11-04 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1964886.42401 | 0.000247 | [] | | 675 | 2014-11-05 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2049228.79401 | -0.003869 | [] | | 676 | 2014-11-06 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2020489.70401 | 0.001047 | [] | | 677 | 2014-11-07 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2027362.02401 | -0.001564 | [] | | 678 | 2014-11-10 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2043606.06401 | 0.025410 | [] | | 679 | 2014-11-11 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2022988.64401 | -0.002775 | [] | | 680 | 2014-11-12 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2049228.91401 | 0.013957 | [] | | 681 | 2014-11-13 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2054226.61401 | -0.005616 | [] | | 682 | 2014-11-14 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1987377.18401 | 0.000519 | [] | | 683 | 2014-11-17 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1988626.85401 | -0.005420 | [] | | 684 | 2014-11-18 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2006744.97401 | -0.010004 | [] | | 685 | 2014-11-19 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2014241.95401 | -0.001653 | [] | | 686 | 2014-11-20 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1980504.81401 | -0.000047 | [] | | 687 | 2014-11-21 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 1989251.55401 | 0.018273 | [] | | 688 | 2014-11-24 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2085464.99401 | 0.025470 | [] | | 689 | 2014-11-25 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2156687.78401 | 0.013702 | [] | | 690 | 2014-11-26 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2142942.92401 | 0.013949 | [] | | 691 | 2014-11-27 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2146691.26401 | 0.011557 | [] | | 692 | 2014-11-28 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2276016.94401 | 0.019724 | [] | | 693 | 2014-12-01 | 1.56401 | {u'000625.XSHE': 1.0, u'601633.XSHG': 62476.0} | 2245404.03401 | 0.003913 | [] | ``` 694 rows × 6 columns ``` ```py perf = qp.perf_parse(bt) out_keys = ['annualized_return', 'volatility', 'information', 'sharpe', 'max_drawdown', 'alpha', 'beta'] for k in out_keys: print '%s: %s' % (k, perf[k]) annualized_return: 0.448632577093 volatility: 0.397466535866 information: 0.825863671828 sharpe: 1.04326663926 max_drawdown: 0.518092986656 alpha: 0.392363999248 beta: 0.886220585368 ``` ```py perf['cumulative_return'].plot() perf['benchmark_cumulative_return'].plot() pylab.legend(['current_strategy','HS300']) <matplotlib.legend.Legend at 0x4e27c50> ``` ![](https://box.kancloud.cn/2016-07-30_579cbdb352cd9.png)