程式交易策略分享，利用ChatGPT協助來編寫Python程式碼

以下是我請chatGPT產生幾個常見的程式交易策略，並附上用Python的語法寫出這些策略的基本概念。請注意，這些僅作為範例，並非實際交易建議。

1. 移動平均線交叉策略

這是一個基於移動平均線的策略，當短期移動平均線穿越長期移動平均線時進行買賣。

import pandas as pd



def moving_average_crossover(data, short_window, long_window):

    signals = pd.DataFrame(index=data.index)

    signals['signal'] = 0.0



    # 短期和長期移動平均線

    signals['short_mavg'] = data['Close'].rolling(window=short_window, min_periods=1, center=False).mean()

    signals['long_mavg'] = data['Close'].rolling(window=long_window, min_periods=1, center=False).mean()



    # 產生交易信號

    signals['signal'][short_window:] = np.where(signals['short_mavg'][short_window:] > signals['long_mavg'][short_window:], 1.0, 0.0)   



    # 產生交易指令

    signals['positions'] = signals['signal'].diff()

    

    return signals

2. 均線回歸策略

當價格偏離其歷史均值時，進行買入或賣出，期望價格回歸均值。

def mean_reversion_strategy(data, window, std_dev):

    mean = data['Close'].rolling(window=window).mean()

    std = data['Close'].rolling(window=window).std()

    

    data['upper_band'] = mean + (std_dev * std)

    data['lower_band'] = mean - (std_dev * std)



    # 買入信號

    data['long_entry'] = data['Close'] < data['lower_band']

    

    # 賣出信號

    data['short_entry'] = data['Close'] > data['upper_band']



    return data

3. 相對強弱指數(RSI)策略

利用RSI指標來判斷市場是超買還是超賣。

def RSI_strategy(data, periods=14, upper_bound=70, lower_bound=30):

    delta = data['Close'].diff()

    gain = (delta.where(delta > 0, 0)).rolling(window=periods).mean()

    loss = (-delta.where(delta < 0, 0)).rolling(window=periods).mean()



    RS = gain / loss

    RSI = 100 - (100 / (1 + RS))



    data['RSI'] = RSI

    data['buy_signal'] = np.where(data['RSI'] < lower_bound, 1, 0)

    data['sell_signal'] = np.where(data['RSI'] > upper_bound, -1, 0)



    return data

4. 布林帶策略

利用布林帶來識別市場的過度買入或過度賣出狀態。

def bollinger_band_strategy(data, window, num_std_dev):

    rolling_mean = data['Close'].rolling(window=window).mean()

    rolling_std = data['Close'].rolling(window=window).std()



    data['upper_band'] = rolling_mean + (rolling_std * num_std_dev)

    data['lower_band'] = rolling_mean - (rolling_std * num_std_dev)



    data['long_entry'] = data['Close'] < data['lower_band']

    data['short_entry'] = data['Close'] > data['upper_band']

    

    return data

5. 交易量加權平均價格(VWAP)策略

VWAP是一種交易算法，用於計算一個資產的平均價格，基於交易量和價格。

def vwap_strategy(data):

    vwap = (data['Volume'] * data['Close']).cumsum() / data['Volume'].cumsum()

    data['VWAP'] = vwap

    # 策略信號可根據VWAP和價格關係定義

    return data

6. 動量交易策略

根據資產的歷史價格變動來評估其未來運動的策略。

def momentum_strategy(data, window):

    data['momentum'] = data['Close'] - data['Close'].shift(window)

    data['signal'] = np.where(data['momentum'] > 0, 1, -1)

    return data

7. 套利策略

尋找並利用兩個相關資產之間的價格差異。

def pairs_trading_strategy(data, asset1, asset2):

    spread = data[asset1] - data[asset2]

    mean_spread = spread.mean()

    std_spread = spread.std()



    data['spread'] = spread

    data['z_score'] = (spread - mean_spread) / std_spread



    data['long_entry'] = data['z_score'] < -2  # 買入標準差低值

    data['short_entry'] = data['z_score'] > 2  # 賣出標準差高值

    

    return data

總結

這些策略都是基於Python的Pandas庫，用於對股票市場數據進行操作和分析。在實際應用這些策略之前，建議進行徹底程式交易的回測和風險評估。每個策略的效果會根據市場條件、選擇的參數和其他因素而有所不同。