Mean-Reversion Trading of Methanol and Polymer Production Spreads
Summary
This example describes a futures spread strategy built around an estimated methanol-to-olefins production margin. It combines the values of polyethylene and polypropylene output with methanol input costs, scaling prices by contract multipliers and assumed production ratios. A rolling history of daily spreads supplies a mean and standard deviation; the current spread’s z-score is used to identify deviations.
The strategy takes opposite positions across the three contracts when the z-score crosses either side of a threshold, aiming to profit if the spread returns toward its historical mean. It closes when the score nears zero and includes a further-deviation stop condition. The document gives example contract choices, a lookback, thresholds, and a backtest interval, but reports no performance results. The production ratios are assumptions, and the code does not account for factors such as fees, slippage, changing contract relationships, or execution risk. Its position-state handling and spread normalization also warrant careful review before drawing conclusions from a backtest.
Key ideas
- The spread estimates output value from polyethylene and polypropylene less the cost of methanol input.
- Historical spread mean and standard deviation are used to calculate a z-score.
- The strategy enters offsetting futures positions when the spread deviates beyond a threshold.
- Positions are closed near the mean, with an additional stop condition for further divergence.
- Assumed production ratios and omitted trading frictions limit the interpretation of the example.
Tags
Full text
# mto_spread
# mto_spread
## Source (Apache-2.0)
```python
#!/usr/bin/env python
# coding=utf-8
__author__ = "Chaos"
from datetime import date
from tqsdk import TqApi, TqAuth, TargetPosTask, TqBacktest, BacktestFinished
import numpy as np
import time
# === 用户参数 ===
# 合约参数
MA = "CZCE.MA409" # 甲醇期货合约
L = "DCE.l2409" # 聚乙烯期货合约
PP = "DCE.pp2409" # 聚丙烯期货合约
START_DATE = date(2023, 11, 1) # 回测开始日期
END_DATE = date(2024, 4, 30) # 回测结束日期
# 套利参数
LOOKBACK_DAYS = 30 # 计算历史价差的回溯天数
STD_THRESHOLD = 2.0 # 标准差阈值，超过此阈值视为套利机会
ORDER_VOLUME = 100 # 聚乙烯的下单手数
CLOSE_THRESHOLD = 0.5 # 平仓阈值(标准差)
# 生产比例（可根据实际工艺调整）
MA_RATIO = 3 # 生产1吨烯烃消耗3吨甲醇
L_RATIO = 1 # 产出1吨聚乙烯
PP_RATIO = 1 # 产出1吨聚丙烯
# === 初始化API ===
api = TqApi(backtest=TqBacktest(start_dt=START_DATE, end_dt=END_DATE),
auth=TqAuth("快期账户", "快期密码"))
# 获取合约行情和K线
ma_quote = api.get_quote(MA)
l_quote = api.get_quote(L)
pp_quote = api.get_quote(PP)
ma_klines = api.get_kline_serial(MA, 60*60*24, LOOKBACK_DAYS)
l_klines = api.get_kline_serial(L, 60*60*24, LOOKBACK_DAYS)
pp_klines = api.get_kline_serial(PP, 60*60*24, LOOKBACK_DAYS)
# 创建目标持仓任务
ma_pos = TargetPosTask(api, MA)
l_pos = TargetPosTask(api, L)
pp_pos = TargetPosTask(api, PP)
# 获取合约乘数
ma_volume_multiple = ma_quote.volume_multiple
l_volume_multiple = l_quote.volume_multiple
pp_volume_multiple = pp_quote.volume_multiple
# 初始化状态变量
position_time = 0 # 建仓时间
in_position = False # 是否有持仓
mean_spread = 0 # 历史价差均值
std_spread = 0 # 历史价差标准差
print(f"策略启动，监控合约: {MA}, {L}, {PP}")
# === 主循环 ===
try:
# 初始计算历史统计值
spreads = []
for i in range(len(ma_klines) - 1):
ma_price = ma_klines.close.iloc[i] * ma_volume_multiple * MA_RATIO
l_price = l_klines.close.iloc[i] * l_volume_multiple * L_RATIO
pp_price = pp_klines.close.iloc[i] * pp_volume_multiple * PP_RATIO
# MTO利润 = (L价值 + PP价值) - MA成本
spread = (l_price + pp_price) - ma_price
spreads.append(spread)
mean_spread = np.mean(spreads)
std_spread = np.std(spreads)
print(f"历史MTO利润均值: {mean_spread:.2f}, 标准差: {std_spread:.2f}")
# 主循环
while True:
api.wait_update()
# 当K线数据有变化时进行计算
if api.is_changing(ma_klines) or api.is_changing(l_klines) or api.is_changing(pp_klines):
# 重新计算历史价差统计
spreads = []
for i in range(len(ma_klines) - 1):
ma_price = ma_klines.close.iloc[i] * ma_volume_multiple * MA_RATIO
l_price = l_klines.close.iloc[i] * l_volume_multiple * L_RATIO
pp_price = pp_klines.close.iloc[i] * pp_volume_multiple * PP_RATIO
spread = (l_price + pp_price) - ma_price
spreads.append(spread)
mean_spread = np.mean(spreads)
std_spread = np.std(spreads)
# 计算当前利润价差
ma_price = ma_klines.close.iloc[-1] * ma_volume_multiple * MA_RATIO
l_price = l_klines.close.iloc[-1] * l_volume_multiple * L_RATIO
pp_price = pp_klines.close.iloc[-1] * pp_volume_multiple * PP_RATIO
current_spread = (l_price + pp_price) - ma_price
# 计算z-score (标准化的价差)
z_score = (current_spread - mean_spread) / std_spread
print(f"当前MTO利润: {current_spread:.2f}, Z-score: {z_score:.2f}")
# 获取当前持仓
ma_position = api.get_position(MA)
l_position = api.get_position(L)
pp_position = api.get_position(PP)
current_ma_pos = ma_position.pos_long - ma_position.pos_short
current_l_pos = l_position.pos_long - l_position.pos_short
current_pp_pos = pp_position.pos_long - pp_position.pos_short
# 计算实际下单手数（依据比例）
ma_volume = int(ORDER_VOLUME * MA_RATIO / L_RATIO)
# L和PP按同等手数下单
# === 交易信号判断 ===
if not in_position: # 如果没有持仓
if z_score > STD_THRESHOLD: # 利润显著高于均值
# 做空利润：卖出L和PP，买入MA
print(f"做空利润：卖出L{ORDER_VOLUME}手和PP{ORDER_VOLUME}手，买入MA{ma_volume}手")
l_pos.set_target_volume(-ORDER_VOLUME)
pp_pos.set_target_volume(-ORDER_VOLUME)
ma_pos.set_target_volume(ma_volume)
position_time = time.time()
in_position = True
elif z_score < -STD_THRESHOLD: # 利润显著低于均值
# 做多利润：买入L和PP，卖出MA
print(f"做多利润：买入L{ORDER_VOLUME}手和PP{ORDER_VOLUME}手，卖出MA{ma_volume}手")
l_pos.set_target_volume(ORDER_VOLUME)
pp_pos.set_target_volume(ORDER_VOLUME)
ma_pos.set_target_volume(-ma_volume)
position_time = time.time()
in_position = True
else: # 如果已有持仓
# 检查是否应当平仓
if abs(z_score) < CLOSE_THRESHOLD: # 利润回归正常
print("利润回归正常，平仓所有头寸")
l_pos.set_target_volume(0)
pp_pos.set_target_volume(0)
ma_pos.set_target_volume(0)
in_position = False
# 止损逻辑
if (z_score > STD_THRESHOLD * 1.5 and current_l_pos < 0) or \
(z_score < -STD_THRESHOLD * 1.5 and current_l_pos > 0):
print("止损：利润向不利方向进一步偏离")
l_pos.set_target_volume(0)
pp_pos.set_target_volume(0)
ma_pos.set_target_volume(0)
in_position = False
except BacktestFinished as e:
print("回测结束")
api.close()
```Shown in full with attribution under the source's licence. Licence: Apache-2.0
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.