Pharos
TAMomentum

TA.MACD()

移动平均收敛散度 (Moving Average Convergence/Divergence)

通过快慢 EMA 的差值判断趋势和动能，最常用的趋势跟踪指标之一。

语法

python
TA.MACD(data, fastperiod=12, slowperiod=26, signalperiod=9)

参数

参数名类型必选默认值说明
dataany否-价格数据数组（通常使用收盘价）
fastperiodany否-快线周期，默认 12
slowperiodany否-慢线周期，默认 26
signalperiodany否-信号线周期，默认 9

返回值

返回 (macd, signal, histogram) 元组：

  • macd: MACD 线（DIF）= EMA(12) - EMA(26)
  • signal: 信号线（DEA）= EMA(MACD, 9)
  • histogram: 柱状图（MACD - Signal）

计算方法

  1. DIF (MACD线) = EMA(Close, 12) - EMA(Close, 26)
  2. DEA (信号线) = EMA(DIF, 9)
  3. MACD柱 = DIF - DEA

信号解读

金叉死叉

  • 金叉: MACD 上穿 Signal，买入信号
  • 死叉: MACD 下穿 Signal，卖出信号

零轴穿越

  • 上穿零轴: 中长期转为多头
  • 下穿零轴: 中长期转为空头

柱状图

  • 柱状图放大: 动能增强
  • 柱状图缩小: 动能减弱
  • 柱状图变色: 趋势可能反转

背离

  • 顶背离: 价格创新高但 MACD 未创新高，可能反转
  • 底背离: 价格创新低但 MACD 未创新低，可能反弹

基础示例

python
def main():
    records = exchange.GetRecords()
    closes = [r['Close'] for r in records]
    
    macd, signal, hist = TA.MACD(closes, 12, 26, 9)
    
    # MACD 金叉
    if macd[-1] > signal[-1] and macd[-2] <= signal[-2]:
        Log("MACD 金叉，买入信号")
        
    # MACD 死叉
    elif macd[-1] < signal[-1] and macd[-2] >= signal[-2]:
        Log("MACD 死叉，卖出信号")
    
    # 柱状图变化
    if hist[-1] > 0 and hist[-1] > hist[-2]:
        Log("MACD 柱状图放大，多头动能增强")
    elif hist[-1] < 0 and hist[-1] < hist[-2]:
        Log("MACD 柱状图放大，空头动能增强")

高级应用

1. MACD 零轴策略

python
def main():
    records = exchange.GetRecords()
    closes = [r['Close'] for r in records]
    
    macd, signal, hist = TA.MACD(closes, 12, 26, 9)
    
    # 零轴上方金叉（强势）
    if macd[-1] > 0 and macd[-1] > signal[-1] and macd[-2] <= signal[-2]:
        Log("零轴上方金叉，强势买入")
    
    # 零轴下方死叉（弱势）
    elif macd[-1] < 0 and macd[-1] < signal[-1] and macd[-2] >= signal[-2]:
        Log("零轴下方死叉，弱势卖出")
    
    # 穿越零轴
    if macd[-1] > 0 and macd[-2] <= 0:
        Log("MACD 上穿零轴，中期转多")
    elif macd[-1] < 0 and macd[-2] >= 0:
        Log("MACD 下穿零轴，中期转空")

2. MACD 背离检测

python
def detect_macd_divergence(prices, macd_values, period=10):
    """检测 MACD 背离"""
    if len(prices) < period or len(macd_values) < period:
        return None
    
    # 顶背离
    if (prices[-1] > max(prices[-period:-1]) and 
        macd_values[-1] < max(macd_values[-period:-1])):
        return "顶背离"
    
    # 底背离
    if (prices[-1] < min(prices[-period:-1]) and 
        macd_values[-1] > min(macd_values[-period:-1])):
        return "底背离"
    
    return None

def main():
    records = exchange.GetRecords()
    closes = [r['Close'] for r in records]
    
    macd, signal, hist = TA.MACD(closes, 12, 26, 9)
    divergence = detect_macd_divergence(closes, macd, 10)
    
    if divergence == "顶背离":
        Log("MACD 顶背离，价格可能回调")
    elif divergence == "底背离":
        Log("MACD 底背离，价格可能反弹")

3. MACD 柱状图策略

python
def main():
    records = exchange.GetRecords()
    closes = [r['Close'] for r in records]
    
    macd, signal, hist = TA.MACD(closes, 12, 26, 9)
    
    # 柱状图由负转正
    if hist[-1] > 0 and hist[-2] <= 0:
        Log("柱状图由负转正，短期买入信号")
    
    # 柱状图由正转负
    elif hist[-1] < 0 and hist[-2] >= 0:
        Log("柱状图由正转负，短期卖出信号")
    
    # 柱状图连续放大
    if hist[-1] > hist[-2] > hist[-3] and hist[-1] > 0:
        Log("柱状图连续放大，多头动能强")

4. 多周期 MACD 共振

python
def main():
    # 不同周期的K线
    records_1h = exchange.GetRecords(PERIOD_H1)
    records_4h = exchange.GetRecords(PERIOD_H4)
    records_1d = exchange.GetRecords(PERIOD_D1)
    
    # 计算不同周期的 MACD
    macd_1h, signal_1h, _ = TA.MACD([r['Close'] for r in records_1h], 12, 26, 9)
    macd_4h, signal_4h, _ = TA.MACD([r['Close'] for r in records_4h], 12, 26, 9)
    macd_1d, signal_1d, _ = TA.MACD([r['Close'] for r in records_1d], 12, 26, 9)
    
    # 多周期金叉共振
    if (macd_1h[-1] > signal_1h[-1] and 
        macd_4h[-1] > signal_4h[-1] and 
        macd_1d[-1] > signal_1d[-1]):
        Log("多周期MACD金叉共振，强买入信号")

参数优化建议

交易周期推荐参数特点
超短线(6, 13, 5)反应快，信号多
短线(8, 17, 9)较灵敏
标准(12, 26, 9)经典配置
中长线(19, 39, 9)平滑，假信号少

与其他指标配合

MACD + RSI

python
macd, signal, hist = TA.MACD(closes, 12, 26, 9)
rsi = TA.RSI(closes, 14)

# 双重确认
if macd[-1] > signal[-1] and rsi[-1] < 30:
    Log("MACD金叉 + RSI超卖，强买入")

MACD + ADX

python
macd, signal, hist = TA.MACD(closes, 12, 26, 9)
adx = TA.ADX(highs, lows, closes, 14)

# 趋势 + 强度确认
if macd[-1] > signal[-1] and adx[-1] > 25:
    Log("MACD金叉 + 强趋势，高质量信号")

MACD + BOLL

python
macd, signal, hist = TA.MACD(closes, 12, 26, 9)
upper, middle, lower = TA.BOLL(closes, 20, 2)

# 金叉 + 突破中轨
if macd[-1] > signal[-1] and closes[-1] > middle[-1]:
    Log("MACD金叉 + 突破布林中轨")

注意事项

⚠️ 重要提醒：

  1. 震荡市场: MACD 容易产生假信号，建议配合 ADX 使用
  2. 滞后性: MACD 是滞后指标，适合趋势确认而非预测
  3. 参数敏感: 不同周期需要调整参数
  4. 背离信号: 背离是强反转信号，需重点关注
  5. 柱状图: 柱状图变化往往领先于金叉死叉

相关指标

返回目录

← 返回动量指标目录

动量指标综合策略示例

Previous Page

TA.MFI()

Next Page

On this page

TA.MACD()语法参数返回值计算方法信号解读金叉死叉零轴穿越柱状图背离基础示例高级应用1. MACD 零轴策略2. MACD 背离检测3. MACD 柱状图策略4. 多周期 MACD 共振参数优化建议与其他指标配合MACD + RSIMACD + ADXMACD + BOLL注意事项相关指标返回目录