基于ricequant神经网络量化交易

ricequant研究平台代码，用于进行训练模型的生成，训练文件保存

import pandas as pd
import numpy as np# 导入BP模型
from sklearn.neural_network import MLPClassifier
# 导入训练集分割方法
from sklearn.model_selection import train_test_split # 确定每月日期 2014-01-01~2016-01-01
dates = get_trading_dates(start_date="2014-01-01", end_date="2016-01-01")
# 每天日期---->每月月末
# 每月最后一个交易日, 按月计算收益率
month_date = []
for i in range(len(dates) -1):if dates[i].year != dates[i+1].year:month_date.append(dates[i])elif dates[i].month != dates[i+1].month:month_date.append(dates[i])#把最后一个交易日加入
month_date.append(dates[-1])stocks = index_components("000300.XSHG")all_data = pd.DataFrame()
len_date = len(month_date[:-1])
for i in range(0, len_date):date = month_date[i]date_next = Noneif i + 1 < len_date:date_next = month_date[i+1]fund = get_factor(stocks, ["pe_ratio_lyr","pb_ratio_lyr","market_cap","ev_lyr","return_on_asset_net_profit_lyr","du_return_on_equity_lyr","basic_earnings_per_share","net_profit_to_revenue_lyr","total_expense"],start_date=date,end_date=date)fund.reset_index(inplace=True)fund.sort_index(inplace=True)fund['next_month_return'] = 0if not (date_next is None):for i in range(0, fund.shape[0]):stock = fund.iloc[i]["order_book_id"]price1 = get_price(stock, start_date=date, end_date=date, fields='close')price2 = get_price(stock, start_date=date_next, end_date=date_next, fields='close')if (price1 is not None) and (price2 is not None):price1.reset_index(inplace=True)price2.reset_index(inplace=True)#fund.loc[i:i,('next_month_return')] = (float)(price2["close"] / price1["close"]) - 1 #赋值不强转就赋值不上if (float)(price2["close"] / price1["close"]) - 1 > 0.1:fund.loc[i:i,('next_month_return')] = 1# 进行每月因子数据拼接all_data = pd.concat([all_data, fund])# 把收益率为空删除
all_data = all_data.dropna()def mad(factor):"""3倍中位数去极值"""# 求出因子值的中位数med = np.median(factor)# 求出因子值与中位数的差值，进行绝对值mad = np.median(np.abs(factor - med))# 定义几倍的中位数上下限high = med + (3 * 1.4826 * mad)low = med - (3 * 1.4826 * mad)# 替换上下限以外的值factor = np.where(factor > high, high, factor)factor = np.where(factor < low, low, factor)return factordef stand(factor):"""标准化"""mean = np.mean(factor)std = np.std(factor)return (factor - mean)/std#随机打乱，并采样，index已乱
all_data = all_data.sample(frac=0.7) #训练数据
x = all_data[["pe_ratio_lyr","pb_ratio_lyr","market_cap","ev_lyr","return_on_asset_net_profit_lyr","du_return_on_equity_lyr","basic_earnings_per_share","net_profit_to_revenue_lyr","total_expense"]].copy()# 取出目标值
y = all_data[['next_month_return']].copy()# 1、特征值处理
# 去极值、标准化、中性化
for name in x.columns:x[name] = mad(x[name])x[name] = stand(x[name])#分为训练集与测试集
x_train,x_test = train_test_split(x,test_size=0.1,random_state=42)#训练集
y_train,y_test = train_test_split(y,test_size=0.1,random_state=42)#测试集# 建立 BP 模型, 采用Adam优化器，relu非线性映射函数
BP = MLPClassifier(solver='adam',activation = 'relu',max_iter = 10000,alpha = 1e-3,hidden_layer_sizes = (64,32, 32),random_state = 1)
# 进行模型训练
BP.fit(x_train.values, y_train.values.ravel())# 进行模型预测
#predict_train_labels = BP.predict(x_test)y_predict = BP.predict(x_train)#精确率
acc = np.mean([(x == y)  for x, y in zip(y_predict,y_train.values) if y==1])
accimport pickle# 保存模型
with open('model.txt', 'wb') as f:pickle.dump(BP, f)# 读取模型
with open('model.txt', 'rb') as f:model = pickle.load(f)
model.predict(x_train)

我的策略进行模型效果分析

import pickle
import pandas as pd
import numpy as np# 导入BP模型
from sklearn.neural_network import MLPClassifier
# 导入训练集分割方法
from sklearn.model_selection import train_test_split
from six import StringIO
from six import BytesIO
# 在这个方法中编写任何的初始化逻辑。context对象将会在你的算法策略的任何方法之间做传递。
def init(context):body = get_file('model.txt')context.model = pickle.load(BytesIO(body))#model.predict(x_train)context.hs300 = index_components("000300.XSHG")scheduler.run_monthly(regression_select, tradingday=1)def regression_select(context, bar_dict):fund = get_factor(context.hs300, ["pe_ratio_lyr","pb_ratio_lyr","market_cap","ev_lyr","return_on_asset_net_profit_lyr","du_return_on_equity_lyr","basic_earnings_per_share","net_profit_to_revenue_lyr","total_expense"])fund.dropna(inplace=True)fund.reset_index(inplace=True)#训练数据x = fund[["pe_ratio_lyr","pb_ratio_lyr","market_cap","ev_lyr","return_on_asset_net_profit_lyr","du_return_on_equity_lyr","basic_earnings_per_share","net_profit_to_revenue_lyr","total_expense"]].copy()# 1、特征值处理# 去极值、标准化、中性化for name in x.columns:x[name] = mad(x[name])x[name] = stand(x[name])y = context.model.predict(x)stock_return = dict(zip(fund["order_book_id"], y))# 对字典进行排序score = sorted(stock_return.items(), key=lambda x: x[1], reverse=True)[:20]# 取出score的股票代码context.stocklist = [x[0] for x in score]rebalance(context)def rebalance(context):# 卖出for stock in context.portfolio.positions.keys():if context.portfolio.positions[stock].quantity > 0:if stock not in context.stocklist:order_target_percent(stock, 0)weight = 1.0 / len(context.stocklist)# 买入for stock in context.stocklist:order_target_percent(stock, weight)def rebalance(context):# 卖出for stock in context.portfolio.positions.keys():if context.portfolio.positions[stock].quantity > 0:if stock not in context.stocklist:order_target_percent(stock, 0)weight = 1.0 / len(context.stocklist)# 买入for stock in context.stocklist:order_target_percent(stock, weight)# before_trading此函数会在每天策略交易开始前被调用，当天只会被调用一次
def before_trading(context):pass# 你选择的证券的数据更新将会触发此段逻辑，例如日或分钟历史数据切片或者是实时数据切片更新
def handle_bar(context, bar_dict):pass# after_trading函数会在每天交易结束后被调用，当天只会被调用一次
def after_trading(context):passdef mad(factor):"""3倍中位数去极值"""# 求出因子值的中位数med = np.median(factor)# 求出因子值与中位数的差值，进行绝对值mad = np.median(np.abs(factor - med))# 定义几倍的中位数上下限high = med + (3 * 1.4826 * mad)low = med - (3 * 1.4826 * mad)# 替换上下限以外的值factor = np.where(factor > high, high, factor)factor = np.where(factor < low, low, factor)return factordef stand(factor):"""标准化"""mean = np.mean(factor)std = np.std(factor)return (factor - mean)/std

交易效果图

通过此量化交易策略，获得了超出市场一倍的超额收益。

基于ricequant神经网络量化交易相关推荐

python 量化交易_基于Python的量化交易工具清单(上）
-- Python量化工具清单 -- 以下内容来源于Wilson Freitas的Github项目"Awesome Quant".原文中包含了丰富的语言类别,但是后续介绍主要针对P ...
R语言神经网络量化交易模型
上篇是逻辑回归模型,这次我们用神经网络模型.再算一遍,试试看 # 载入示例股票library(quantmod) getSymbols("^DJI", src = "ya ...
基于人性的量化交易：期货的反向跟单交易软件系统
一. 反向跟单的概念一.财智汇反向跟单系统简介一款不用客户手动操作,利用逆向思维跟踪小白亏货操作的自动交易软件. 转载于:https://my.oschina.net/u/617911/blog/ ...
基于Informer的股价预测（量化交易综述）
摘要股票市场是金融市场中不可或缺的组成部分.准确预测股票趋势对于投资者和市场参与者具有重要意义,因为它们可以指导投资决策.优化投资组合以及降低金融风险.而且可以提升国家国际地位以及金融风险控制能力, ...
基于人工智能的期权量化交易
基于人工智能的期权量化交易基于人工智能的期权量化交易基于人工智能的期权量化交易该文基于人工智能AI的深度强化学习,进行股票期权的量化投资策略研究及回测评估.作者建立了人工智能学习及交易系统.基于 ...
python量化投资必背代码-基于python的开源量化交易，量化投资架构
原标题:基于python的开源量化交易,量化投资架构 github地址:https://github.com/bbfamily/abu abu能够帮助用户自动完善策略,主动分析策略产生的交易行为,智能 ...
python 量化交易框架开源_Hikyuu首页、文档和下载 - 基于 C++/Python 的开源量化交易研究框架 - OSCHINA - 中文开源技术交流社区...
Hikyuu Quant Framework是一款基于C++/Python的开源量化交易研究框架,用于策略分析及回测.其核心思想基于当前成熟的系统化交易方法,将整个系统化交易抽象为由市场环境判断策略. ...
python量化交易开源框架_hikyuu: 基于C++/Python的开源量化交易研究框架
Hikyuu Quant Framework是一款基于C++/Python的开源量化交易研究框架,用于策略分析及回测(仅受限于数据,如有数据也可用于期货等).其核心思想基于当前成熟的系统化交易方法,将 ...
python统计套利_清华编程高手尹成带你基于算法实践python量化交易
清华编程高手尹成带你基于算法实践python量化交易量化交易是指以先进的数学模型替代人为的主观判断,利用计算机技术从庞大的历史数据中海选能带来超额收益的多种"大概率"事件以制定策 ...
python量化交易书籍推荐知乎_GitHub - XingkaiLiang/vnpy: 基于python的开源量化交易平台开发框架...
vn.py - By Traders, For Traders. 简介 vn.py是基于Python的开源量化交易程序开发框架,起源于国内私募的自主量化交易系统.2015年初项目启动时只是单纯的交易A ...

基于ricequant神经网络量化交易

基于ricequant神经网络量化交易相关推荐

最新文章

热门文章