import numpy as np import pandas as pd import talib def initialize(context): context.symbol = 'BTCUSDT' context.window_size = 5 context.deviation = 1 context.trade_size = 0.01 context.stop_loss = 0.05 context.take_profit = 0.1 schedule_function(rebalance, date_rules.every_day(), time_rules.market_open()) def rebalance(context, data): price = data.history(context.symbol, 'close', context.window_size + 1, '1d') signal = mean_reversion_signal(price, context.window_size, context.deviation) current_position = context.portfolio.positions[context.symbol].amount if signal[-1] == 1 and current_position <= 0: target_position_size = context.trade_size / data.current(context.symbol, 'close') order_target_percent(context.symbol, target_position_size) elif signal[-1] == -1 and current_position >= 0: order_target(context.symbol, 0) elif current_position > 0: current_price = data.current(context.symbol, 'close') stop_loss_price = current_price * (1 - context.stop_loss) take_profit_price = current_price * (1 + context.take_profit) if current_price <= stop_loss_price or current_price >= take_profit_price: order_target(context.symbol, 0) def moving_average(x, n): ma = talib.SMA(x, timeperiod=n) return ma def std_deviation(x, n): std = talib.STDDEV(x, timeperiod=n) return std def mean_reversion_signal(price, window_size, deviation): ma = moving_average(price, window_size) std = std_deviation(price, window_size) upper_band = ma + deviation * std lower_band = ma - deviation * std signal = np.zeros_like(price) signal[price > upper_band] = -1 # 卖出信号 signal[price < lower_band] = 1 # 买入信号 return signal ''' 运行回测 ''' start_date = pd.to_datetime('2019-01-01', utc=True) end_date = pd.to_datetime('2021-01-01', utc=True) results = run_algorithm( start=start_date, end=end_date, initialize=initialize, capital_base=10000, data_frequency='daily', bundle='binance' ) ''' 查看回测结果 ''' print(results.portfolio_value)格式错误

这段代码存在一些格式错误。具体来说，第一行和第二行之间需要有一个空行，否则会提示语法错误。另外，代码中的注释也需要使用 # 符号进行标记。

以下是修改后的代码：

import numpy as np
import pandas as pd
import talib

def initialize(context):
    context.symbol = 'BTCUSDT'
    context.window_size = 5
    context.deviation = 1
    context.trade_size = 0.01
    context.stop_loss = 0.05
    context.take_profit = 0.1
    
    schedule_function(rebalance, date_rules.every_day(), time_rules.market_open())

def rebalance(context, data):
    price = data.history(context.symbol, 'close', context.window_size + 1, '1d')
    signal = mean_reversion_signal(price, context.window_size, context.deviation)
    current_position = context.portfolio.positions[context.symbol].amount
    
    if signal[-1] == 1 and current_position <= 0:
        target_position_size = context.trade_size / data.current(context.symbol, 'close')
        order_target_percent(context.symbol, target_position_size)
    elif signal[-1] == -1 and current_position >= 0:
        order_target(context.symbol, 0)
    elif current_position > 0:
        current_price = data.current(context.symbol, 'close')
        stop_loss_price = current_price * (1 - context.stop_loss)
        take_profit_price = current_price * (1 + context.take_profit)
        
        if current_price <= stop_loss_price or current_price >= take_profit_price:
            order_target(context.symbol, 0)

def moving_average(x, n):
    ma = talib.SMA(x, timeperiod=n)
    return ma

def std_deviation(x, n):
    std = talib.STDDEV(x, timeperiod=n)
    return std

def mean_reversion_signal(price, window_size, deviation):
    ma = moving_average(price, window_size)
    std = std_deviation(price, window_size)
    upper_band = ma + deviation * std
    lower_band = ma - deviation * std
    
    signal = np.zeros_like(price)
    signal[price > upper_band] = -1  # 卖出信号
    signal[price < lower_band] = 1   # 买入信号
    
    return signal

''' 运行回测 '''
start_date = pd.to_datetime('2019-01-01', utc=True)
end_date = pd.to_datetime('2021-01-01', utc=True)

results = run_algorithm(
    start=start_date,
    end=end_date,
    initialize=initialize,
    capital_base=10000,
    data_frequency='daily',
    bundle='binance'
)

''' 查看回测结果 '''
print(results.portfolio_value)