Name
基于双MACD量化交易策略Dual-MACD-Quantitative-Trading-Strategy
Author
ChaoZhang
Strategy Description
双MACD量化交易策略是一种利用双时间框架MACD指标实现的量化交易策略。该策略在周线MACD指标形成金叉时开仓做多,在日线MACD指标形成死叉时平仓。当仓位为空时,若日线MACD指标再次形成金叉,则可以重新开仓做多。
双MACD量化交易策略利用周MACD指标和日MACD指标的组合来判断入场和出场信号。
首先,当周MACD指标的MACD线上穿信号线时产生买入信号,此时开仓做多;然后当日MACD指标的MACD线下穿信号线时产生卖出信号,此时平仓。
当仓位为空时,若日MACD指标的MACD线再次上穿信号线,则重新开仓做多。也就是说,日MACD指标的金叉成为再次开仓的条件。
需要注意的是,日MACD指标的死叉才会平仓,但必须在周MACD指标MACD线高于信号线的“交易窗口”内才允许再次开仓。
双MACD量化交易策略结合了双时间框架分析,可以有效过滤假信号,提高信号的质量。具体来说,主要有以下几个优势:
-
周时间框架判断主要趋势方向,有助于避免逆势交易。
-
日时间框架判断入场和出场时机,可以及时捕捉短期交易机会。
-
“交易窗口”机制可以避免因短期调整而过于频繁开仓平仓。
-
MACD指标参数可调,可以根据不同品种和市场环境进行优化。
-
整合止盈、止损、移动止损功能,可以有效控制风险。
双MACD量化交易策略也存在一定的风险,主要包括:
-
MACD指标容易产生假信号和频繁交叉,需要组合其他指标进行确认。
-
周月时间框架判断的主要趋势可能发生转折,需要及时止损。
-
Parameters需要根据不同品种和行情环境不断优化和调整。
-
不能过度依赖回测结果,实盘可能与回测有差异。
对应解决方法:
-
与其他指标组合使用,构建逻辑优化的策略体系。
-
设置合理的止损幅度,避免超过可承受的最大损失。
-
不断优化参数,寻找最佳参数组合。
-
从最小资金开始实盘,验证策略稳定性。
双MACD量化交易策略还有进一步优化的空间:
-
可以引入布林线、KDJ等其他指标,构建多指标组合策略,提高信号质量。
-
可以结合交易量指标,避免价格上涨但成交量不足的假突破。
-
可以利用机器学习方法自动优化参数,实现参数的动态调整。
-
可以针对策略作进一步风险调整,如加入盈亏比等高级止损方法。
-
策略拟合性检验与优化调整,避免过拟合问题。
双MACD量化交易策略整合双时间框架分析判断主副趋势,以发挥各自指标优势。策略优化空间还很大,有望通过引入其他指标、利用机器学习进行参数优化等方法进一步提升策略效果。实盘验证是必不可少的一步,也是进一步完善策略的重要依据。
||
The Dual MACD quantitative trading strategy is a quantitative trading strategy implemented using dual timeframe MACD indicators. It goes long when the weekly MACD indicator forms a golden cross and closes the position when the daily MACD indicator forms a death cross. When the position is empty, if the daily MACD indicator forms another golden cross, a new long position can be opened.
The Dual MACD quantitative trading strategy uses a combination of weekly MACD and daily MACD indicators to determine entry and exit signals.
Firstly, when the MACD line of the weekly MACD indicator crosses above the signal line, a buy signal is generated and a long position is opened. Then when the MACD line of the daily MACD indicator crosses below the signal line, a sell signal is generated and the position is closed.
When the position is empty, if the MACD line of the daily MACD indicator crosses above the signal line again, a new long position is reopened. That is to say, the golden cross of the daily MACD acts as the re-entry condition.
Note that only the death cross of the daily MACD will close the position, but reopening is only allowed when the MACD line of the weekly MACD is above the signal line, within the "Trading Window".
The Dual MACD quantitative trading strategy combines dual timeframe analysis, which can effectively filter false signals and improve signal quality. Specifically, there are several main advantages:
-
The weekly timeframe judges the main trend direction, which helps avoid contrarian trading.
-
The daily timeframe determines entry and exit timing, which can timely capture short-term trading opportunities.
-
The "Trading Window" mechanism can avoid excessively frequent opening and closing due to short-term adjustments.
-
The MACD indicator parameters are adjustable and can be optimized according to different varieties and market conditions.
-
Integrates take profit, stop loss, trailing stop loss functions to effectively control risks.
The Dual MACD quantitative trading strategy also has some risks, mainly including:
-
MACD indicator tends to generate false signals and frequent crossovers, needs confirmation from other indicators.
-
The main trend identified in weekly/monthly timeframe may reverse, trailing stop loss is necessary.
-
Parameters need continuous optimization and adjustment according to varieties and market conditions.
-
Cannot overly rely on backtest results, live performance may differ from backtest.
Corresponding solutions:
-
Combine with other indicators to build strategy systems with logic optimization.
-
Set reasonable stop loss to avoid exceeding maximum tolerable loss.
-
Continuously optimize parameters to find optimal combinations.
-
Start live trading from minimum capital to validate stability.
The Dual MACD quantitative trading strategy has room for further optimization:
-
Introduce Bollinger Bands, KDJ and other indicators to build multi-indicator combined strategies and improve signal quality.
-
Incorporate trading volume indicators to avoid false breakouts with insufficient volume.
-
Utilize machine learning methods to automatically optimize parameters and achieve dynamic adjustment.
-
Further risk adjustment of the strategy, such as adding advanced stop loss methods like profit & loss ratio.
-
Strategy fitness test & optimization to avoid overfitting problems.
The Dual MACD quantitative trading strategy integrates dual timeframe analysis to determine main and subordinate trends and gives full play to the advantages of each indicator. There is still great potential for strategy optimization, and it is expected to further improve strategy performance by introducing other indicators, automatic parameter optimization through machine learning, etc. Live trading verification is an indispensable step and important basis for further perfecting the strategy.
[/trans]
Strategy Arguments
| Argument | Default | Description |
|---|---|---|
| v_input_1 | timestamp(01 May 2018 13:30 +0000) | Start Time |
| v_input_2 | timestamp(9 Sep 2021 13:30 +0000) | Finish Time |
| v_input_3 | false | ------ Profit & Loss ------ |
| v_input_4 | true | Enable Just a Profit Level? |
| v_input_5 | false | Enable Just a S.Loss Level? |
| v_input_6 | true | Enable Only Trailing Stop |
| v_input_7 | 30 | Take Profit % |
| v_input_8 | true | Stop Loss % |
| v_input_9 | 5 | Trailing Stop % |
| v_input_10_low | 0 | Trailing Stop Source: low |
| v_input_11 | false | ------ Macd Properties ------ |
| v_input_12 | D | Short Term TimeFrame |
| v_input_13 | W | Long Term TimeFrame |
| v_input_14_close | 0 | Source: close |
| v_input_15 | 12 | Fast Length |
| v_input_16 | 26 | Slow Length |
| v_input_17 | 9 | Sign Length |
| v_input_18 | true | Daily or Weekly? |
Source (PineScript)
/*backtest
start: 2023-01-29 00:00:00
end: 2024-01-11 05:20:00
period: 1d
basePeriod: 1h
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/
// This source code is subject to the terms of the Mozilla Public License 2.0 at https://mozilla.org/MPL/2.0/
// © maxits
// Long Position: Weekly Macd line crosses above Signal line
// [Trading Window Macd Line > Signal Line] (Weekly)
// Close Position: Daily Macd Line crosses above Daily Signal line.
// Re Entry Condition: Macd line crosses above Signal line only if [Trading Window MacdLine > Sgnal Line] (Weekly)
//@version=4
strategy("Dual MACD Strategy",
shorttitle="Dual Macd Tester",
overlay=false,
initial_capital=1000,
default_qty_value=20,
default_qty_type=strategy.percent_of_equity,
commission_value=0.1,
pyramiding=0)
// Define user inputs
i_time = input(defval = timestamp("01 May 2018 13:30 +0000"), title = "Start Time", type = input.time) // Starting time for Backtesting
f_time = input(defval = timestamp("9 Sep 2021 13:30 +0000"), title = "Finish Time", type = input.time) // Finishing time for Backtesting
sep1 = input(false, title="------ Profit & Loss ------")
enable_TP = input(true, title="Enable Just a Profit Level?")
enable_SL = input(false, title="Enable Just a S.Loss Level?")
enable_TS = input(true, title=" Enable Only Trailing Stop")
long_TP_Input = input(30.0, title='Take Profit %', type=input.float, minval=0)/100
long_SL_Input = input(1.0, title='Stop Loss %', type=input.float, minval=0)/100
long_TS_Input = input(5.0, title='Trailing Stop %', type=input.float, minval=0)/100
cl_low_Input = input(low, title="Trailing Stop Source")
long_TP = strategy.position_avg_price * (1 + long_TP_Input)
long_SL = strategy.position_avg_price * (1 - long_SL_Input)
long_TS = cl_low_Input * (1 - long_TS_Input)
sep2 = input(false, title="------ Macd Properties ------")
d_res = input(title="Short Term TimeFrame", type=input.resolution, defval="D") // Daily Time Frame
w_res = input(title="Long Term TimeFrame", type=input.resolution, defval="W") // Weekly Time Frame
src = input(close, title="Source") // Indicator Price Source
fast_len = input(title="Fast Length", type=input.integer, defval=12) // Fast MA Length
slow_len = input(title="Slow Length", type=input.integer, defval=26) // Slow MA Length
sign_len = input(title="Sign Length", type=input.integer, defval=9) // Sign MA Length
d_w = input(title="Daily or Weekly?", type=input.bool, defval=true) // Plot Daily or Weekly MACD
// Color Plot for Macd
col_grow_above = #26A69A
col_grow_below = #FFCDD2
col_fall_above = #B2DFDB
col_fall_below = #EF5350
// BG Color
bg_color = color.rgb(127, 232, 34, 75)
// Daily Macd
[d_macdLine, d_singleLine, d_histLine] = security(syminfo.tickerid, d_res, macd(src, fast_len, slow_len, sign_len)) // Funcion Security para poder usar correcta resolución
plot(d_w ? d_macdLine : na, color=color.blue)
plot(d_w ? d_singleLine : na, color=color.orange)
plot(d_w ? d_histLine : na, style=plot.style_columns,
color=(d_histLine>=0 ? (d_histLine[1] < d_histLine ? col_grow_above : col_fall_above) :
(d_histLine[1] < d_histLine ? col_grow_below : col_fall_below)))
// Weekly Macd
[w_macdLine, w_singleLine, w_histLine] = security(syminfo.tickerid, w_res, macd(src, fast_len, slow_len, sign_len)) // Funcion Security para poder usar correcta resolución
plot(d_w ? na : w_macdLine, color=color.blue)
plot(d_w ? na : w_singleLine, color=color.orange)
plot(d_w ? na : w_histLine, style=plot.style_columns,
color=(w_histLine>=0 ? (w_histLine[1] < w_histLine ? col_grow_above : col_fall_above) :
(w_histLine[1] < w_histLine ? col_grow_below : col_fall_below)))
///////////////////////////////// Entry Conditions
inTrade = strategy.position_size != 0 // Posición abierta
notInTrade = strategy.position_size == 0 // Posición Cerrada
start_time = true
trading_window = w_macdLine > w_singleLine // Weekly Macd Signal enables a trading window
bgcolor(trading_window ? bg_color : na)
buy_cond = crossover (w_macdLine, w_singleLine)
sell_cond = crossunder(d_macdLine, d_singleLine)
re_entry_cond = crossover (d_macdLine, d_singleLine) and trading_window
// Entry Exit Conditions
trailing_stop = 0.0 // Code for calculating Long Positions Trailing Stop Loss
trailing_stop := if (strategy.position_size != 0)
stopValue = long_TS
max(trailing_stop[1], stopValue)
else
0
if (buy_cond and notInTrade and start_time)
strategy.entry(id="First Entry", long=strategy.long, comment="First Long")
if (sell_cond and inTrade)
strategy.close(id="First Entry", comment="Close First Long")
if (re_entry_cond and notInTrade and start_time)
strategy.entry(id="Further Entry", long=strategy.long, comment="Further Entry")
if (sell_cond and inTrade)
strategy.close(id="Further Entry", comment="Close First Long")
if enable_TP
if (enable_TS and not enable_SL)
strategy.exit("Long TP & TS FiEn", "First Entry", limit = long_TP, stop = trailing_stop)
strategy.exit("Long TP & TS FuEn", "Further Entry", limit = long_TP, stop = trailing_stop)
else
if (enable_SL and not enable_TS)
strategy.exit("Long TP & TS FiEn", "First Entry", limit = long_TP, stop = long_SL)
strategy.exit("Long TP & TS FuEn", "Further Entry", limit = long_TP, stop = long_SL)
else
strategy.exit("Long TP & TS FiEn", "First Entry", limit = long_TP)
strategy.exit("Long TP & TS FuEn", "Further Entry", limit = long_TP)
else
if not enable_TP
if (enable_TS and not enable_SL)
strategy.exit("Long TP & TS FiEn", "First Entry", stop = trailing_stop)
strategy.exit("Long TP & TS FuEn", "Further Entry", stop = trailing_stop)
else
if (enable_SL and not enable_TS)
strategy.exit("Long TP & TS FiEn", "First Entry", stop = long_SL)
strategy.exit("Long TP & TS FuEn", "Further Entry", stop = long_SL)
plot(enable_TP ? long_TP : na, title="TP Level", color=color.green, style=plot.style_linebr, linewidth=2)
plot(enable_SL ? long_SL : na, title="SL Level", color=color.red, style=plot.style_linebr, linewidth=2)
plot(enable_TS and trailing_stop ? trailing_stop : na, title="TS Level", color=color.red, style=plot.style_linebr, linewidth=2)
Detail
https://www.fmz.com/strategy/440449
Last Modified
2024-01-30 16:43:29