Overall Statistics |
Total Trades 17 Average Win 3.86% Average Loss -1.15% Compounding Annual Return 6.346% Drawdown 7.900% Expectancy 1.180 Net Profit 20.245% Sharpe Ratio 0.839 Loss Rate 50% Win Rate 50% Profit-Loss Ratio 3.36 Alpha 0.002 Beta 3.11 Annual Standard Deviation 0.077 Annual Variance 0.006 Information Ratio 0.579 Tracking Error 0.077 Treynor Ratio 0.021 Total Fees $42.39 |
import numpy as np import tensorflow as tf class TensorFlowNeuralNetwork(QCAlgorithm): def Initialize(self): self.SetStartDate(2015, 1, 1) # Set Start Date self.SetEndDate(2018, 1, 1) # Set End Date self.SetCash(100000) # Set Strategy Cash spy = self.AddEquity("SPY", Resolution.Minute) self.symbols = [spy.Symbol] self.lookback = 30 self.Schedule.On(self.DateRules.Every(DayOfWeek.Monday), self.TimeRules.AfterMarketOpen("SPY", 28), Action(self.NetTrain)) self.Schedule.On(self.DateRules.Every(DayOfWeek.Monday), self.TimeRules.AfterMarketOpen("SPY", 30), Action(self.Trade)) def OnData(self, data): self.data = data def add_layer(self, inputs, in_size, out_size, activation_function=None): # add one more layer and return the output of this layer Weights = tf.Variable(tf.random_normal([in_size, out_size])) biases = tf.Variable(tf.zeros([1, out_size]) + 0.1) Wx_plus_b = tf.matmul(inputs, Weights) + biases if activation_function is None: outputs = Wx_plus_b else: outputs = activation_function(Wx_plus_b) return outputs def NetTrain(self): history = self.History(self.symbols, self.lookback + 1, Resolution.Daily) self.prices_x, self.prices_y = {}, {} self.sell_prices, self.buy_prices = {}, {} for symbol in self.symbols: if not history.empty: self.prices_x[symbol.Value] = list(history.loc[symbol.Value]['open'][:-1]) self.prices_y[symbol.Value] = list(history.loc[symbol.Value]['open'][1:]) for symbol in self.symbols: if symbol.Value in self.prices_x: # create data x_data = np.array(self.prices_x[symbol.Value]).astype(np.float32).reshape((-1,1)) y_data = np.array(self.prices_y[symbol.Value]).astype(np.float32).reshape((-1,1)) # define placeholder for inputs to network xs = tf.placeholder(tf.float32, [None, 1]) ys = tf.placeholder(tf.float32, [None, 1]) # add hidden layer l1 = self.add_layer(xs, 1, 10, activation_function=tf.nn.relu) # add output layer prediction = self.add_layer(l1, 10, 1, activation_function=None) # the error between prediciton and real data loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction), reduction_indices=[1])) train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss) sess = tf.Session() init = tf.global_variables_initializer() sess.run(init) for i in range(200): # training sess.run(train_step, feed_dict={xs: x_data, ys: y_data}) # predict today's price y_pred_final = sess.run(prediction, feed_dict = {xs: y_data})[0][-1] # self.Debug(f'pred price: {y_pred_final}') self.sell_prices[symbol.Value] = y_pred_final - np.std(y_data) self.buy_prices[symbol.Value] = y_pred_final + np.std(y_data) def Trade(self): # Trending strategy for i in self.Portfolio.Values: # liquidate if self.data[i.Symbol.Value].Open < self.sell_prices[i.Symbol.Value] and i.Invested: self.Liquidate(i.Symbol) # buy if self.data[i.Symbol.Value].Open > self.buy_prices[i.Symbol.Value] and not i.Invested: self.SetHoldings(i.Symbol, 1 / len(self.symbols))