Aberration policy

This is a trend-driven strategy to make money.

• A brief history of my life

Many intuitive futures players may be a bit unfamiliar with the Aberration Strategy, a legendary trading system invented by Keith Fitschen in 1986 that achieved annual returns of more than 100%. In 1993, he published the system in the American hedge fund Future Truth hedge fund magazine. It is a medium to long-term trading system that operates flexibly between eight varieties of grain, meat, metals, energy, forex, stock index futures, etc. It is a long-term trading system that tracks trends to make profits. Since it trades in several unrelated or low-related markets at the same time, it makes up for losses in other volatile markets by making a high profit on one or more of the markets. Its trading frequency is generally 3-4 trades per year, 60% of the time held, with an average of 60 days per trade. The strategy is as follows: first, use the 35-day moving average and the standard deviation of the 35-day closing price to derive the three upper-middle and lower tracks, when the closing price of a Bar is too high when it crosses the tracks, the bottom is open when it crosses the tracks, the bottom crosses the mid-tracks and the top crosses the mid-tracks. The idea behind this strategy is simple: use the moving average and standard deviation channels to capture the trend, and then at any time judge whether the channel has been breached. When the trend is gradually weakening, it tends to slow down because the mid-trajectory will slow down first, so when the K line crosses the mid-trajectory in reverse, the trend can be considered to be over and stopped. The strategy was released years earlier and may not be very applicable in today's trading environment, but the idea of trend capture is enduring.

• A brief introduction to the two

  The Abberation trading system was invented by Keith Fitschen in 1986 and published in Future Trust magazine in 1993. Interestingly, Keith was not a born trader, and he served in the United States Air Force for over 20 years, specializing in weapons navigation systems, which have a profound power in processing time-series data. Abberation trading system is characterized by passive trend following, long-term signals, trading 3-4 times a year on a variety, holding more than 60% of the total trading time, and can allocate different number of varieties according to different capital sizes. It makes huge profits by capturing trends by long-line trading, while trading in several unrelated markets, when one variety pulls back, another may gain. Specifically, the Aberation system uses three tracks to trade. First, the arithmetic average MA (close) of the closing price of the variety over the past N days is calculated as the mid-track (MID), and the standard deviation std (close) of the closing price is calculated as a measure of the volatility, and the mid-track MID+m is calculated.std ((close) and MID-mstd ((close) ); do more when the price breaks out of the uptrend and break even when the price goes back into the mid-track; conversely, do more when the price breaks out of the downtrend and break even when the price goes back into the mid-track.

• The Principles of Strategy

  Aberration is also a channel-breaking system, but its up-and-down path is determined by the rate of oscillation.

  1, up and down tracks: Aberration consists of three lines of channels, in which the central orbit is the moving average of a certain cycle ("AveMa"), and the upper and lower trajectories are based on the central orbit plus or minus a certain price standard deviation ("StdValue"), that is, the Braun line as we know it, and the system is constructed very simply and elegantly.

  The specific calculation process is as follows: (1) Calculation of the mean path: AveMa=Average ((Close[1],Length); (2) Calculated standard deviation: StdValue = StandardDev ((Close[1],Length); (3) Computed on track: UpperBand=Avema+StdDevUpStdValue; (StdDevUp is the uplink parameter) (4) Calculation of the downlink: LowerBand=Avema-StdDevDnStdValue; (StdDevDn is the downlink parameter)

  2 Conditions of sale: The price of the stock has been rising since the beginning of the year, and the price of the stock has been falling since the beginning of the year. In the second half of the year, the price of the stock fell below the level of the previous year's stock market.

• Policy source code

  • Strategic frameworks are logically clear and highly reusable
  • Real-time debugging when running on interactive disks
  • It's running smoothly, and the details are perfect.
  • Multi-variety simultaneous operation, with individual control of the opening volume
  • Automatically restore progress based on position when restarted
  • With wind control modules, real-time risk detection and stop loss
  • Weight Loss Notification is available here
  • If you don't want to rent a server, you can use your own computer or a treehouse to run Windows, Linux, or Mac systems, and the router can be brushed off.

function Aberration(q, e, symbol, period, upRatio, downRatio, opAmount) {
    var self = {}
    self.q = q
    self.e = e
    self.symbol = symbol
    self.upTrack = 0
    self.middleTrack = 0
    self.downTrack = 0
    self.nPeriod = period
    self.upRatio = upRatio
    self.downRatio = downRatio
    self.opAmount = opAmount
    self.marketPosition = 0

    self.lastErrMsg = ''
    self.lastErrTime = ''
    self.lastBar = {
        Time: 0,
        Open: 0,
        High: 0,
        Low: 0,
        Close: 0,
        Volume: 0
    }
    self.symbolDetail = null
    self.lastBarTime = 0
    self.tradeCount = 0
    self.isBusy = false

    self.setLastError = function(errMsg) {
        self.lastErrMsg = errMsg
        self.lastErrTime = errMsg.length > 0 ? _D() : ''
    }

    self.getStatus = function() {
        return [self.symbol, self.opAmount, self.upTrack, self.downTrack, self.middleTrack, _N(self.lastBar.Close), (self.marketPosition == 0 ? "--" : (self.marketPosition > 0 ? "多#ff0000" : "空#0000ff")), self.tradeCount, self.lastErrMsg, self.lastErrTime]
    }
    self.getMarket = function() {
        return [self.symbol, _D(self.lastBarTime), _N(self.lastBar.Open), _N(self.lastBar.High), _N(self.lastBar.Low), _N(self.lastBar.Close), self.lastBar.Volume]
    }

    self.restore = function(positions) {
        for (var i = 0; i < positions.length; i++) {
            if (positions[i].ContractType == self.symbol) {
                self.marketPosition += positions[i].Amount * ((positions[i].Type == PD_LONG || positions[i].Type == PD_LONG_YD) ? 1 : -1)
            }
        }
        if (self.marketPosition !== 0) {
            self.tradeCount++
                Log("恢复", self.symbol, "当前持仓为", self.marketPosition)
        }
    }

    self.poll = function() {
        if (self.isBusy) {
            return false
        }

        if (!$.IsTrading(self.symbol)) {
            self.setLastError("不在交易时间内")
            return false
        }

        if (!self.e.IO("status")) {
            self.setLastError("未连接交易所")
            return false
        }

        var detail = self.e.SetContractType(self.symbol)
        if (!detail) {
            self.setLastError("切换合约失败")
            return false
        }
        if (!self.symbolDetail) {
            self.symbolDetail = detail
            Log("合约", detail.InstrumentName.replace(/\s+/g, ""), ", 策略一次开仓:", self.opAmount, "手, 一手", detail.VolumeMultiple, "份, 最大下单量", detail.MaxLimitOrderVolume, "保证金率:", detail.LongMarginRatio.toFixed(4), detail.ShortMarginRatio.toFixed(4), "交割日期", detail.StartDelivDate);
        }
        var records = self.e.GetRecords()
        if (!records || records.length == 0) {
            self.setLastError("获取柱线失败")
            return false
        }

        var bar = records[records.length - 1]
        self.lastBar = bar

        if (records.length <= self.nPeriod) {
            self.setLastError("柱线长度不够")
            return false
        }

        if (self.lastBarTime < bar.Time) {
            var sum = 0
            var pos = records.length - self.nPeriod - 1
            for (var i = pos; i < records.length - 1; i++) {
                sum += records[i].Close
            }
            var avg = sum / self.nPeriod
            var std = 0
            for (i = pos; i < records.length - 1; i++) {
                std += Math.pow(records[i].Close - avg, 2)
            }
            std = Math.sqrt(std / self.nPeriod)

            self.upTrack = _N(avg + (self.upRatio * std))
            self.downTrack = _N(avg - (self.downRatio * std))
            self.middleTrack = _N(avg)
            self.lastBarTime = bar.Time
        }
        var msg
        var act = ""
        if (self.marketPosition == 0) {
            if (bar.Close > self.upTrack) {
                msg = '做多 触发价: ' + bar.Close + ' 上轨:' + self.upTrack;
                act = "buy"
            } else if (bar.Close < self.downTrack) {
                msg = '做空 触发价: ' + bar.Close + ' 下轨:' + self.downTrack;
                act = "sell"
            }
        } else {
            if (self.marketPosition < 0 && bar.Close > self.middleTrack) {
                msg = '平空 触发价: ' + bar.Close + ' 平仓线:' + self.middleTrack;
                act = "closesell"
            } else if (self.marketPosition > 0 && bar.Close < self.middleTrack) {
                msg = '平多 触发价: ' + bar.Close + ' 平仓线:' + self.middleTrack;
                act = "closebuy"
            }
        }

        if (act == "") {
            return true
        }

        Log(self.symbol + ', ' + msg + (NotifyWX ? '@' : ''))

        self.isBusy = true
        self.tradeCount += 1
        if (self.lastErrMsg != '') {
            self.setLastError('')
        }
        self.q.pushTask(self.e, self.symbol, act, self.opAmount, function(task, ret) {
            self.isBusy = false
            if (!ret) {
                return
            }
            if (task.action == "buy") {
                self.marketPosition = 1
            } else if (task.action == "sell") {
                self.marketPosition = -1
            } else {
                self.marketPosition = 0
            }
        })
    }
    return self
}

function main() {
    if (exchange.GetName() !== 'Futures_CTP') {
        throw "只支持传统商品期货(CTP)"
    }
    
    SetErrorFilter("login|ready|初始化")

    LogStatus("Ready...")
    if (Reset) {
        LogProfitReset()
        LogReset()
    }
    
    // Ref: https://www.fmz.com/bbs-topic/362
    if (typeof(exchange.IO("mode", 0)) == 'number') {
        Log("切换行情模式成功")
    }

    LogStatus("等待与期货商服务器连接..")
    while (!exchange.IO("status")) {
        Sleep(500)
    }
    LogStatus("获取资产信息")
    var tblRuntime = {
        type: 'table',
        title: '交易信息',
        cols: ['品种', '每次开仓量', '上轨', '下轨', '中轨', '最后成交价', '仓位', '交易次数', '最后错误', '错误时间'],
        rows: []
    };
    var tblMarket = {
        type: 'table',
        title: '行情信息',
        cols: ['品种', '当前周期', '开盘', '最高', '最低', '最后成交价', '成交量'],
        rows: []
    };
    var tblPosition = {
        type: 'table',
        title: '持仓信息',
        cols: ['品种', '杠杆', '方向', '均价', '数量', '持仓盈亏'],
        rows: []
    };
    var positions = _C(exchange.GetPosition)
    if (positions.length > 0 && !AutoRestore) {
        throw "程序启动时不能有持仓, 但您可以勾选自动恢复来进行自动识别 !"
    }
    var initAccount = _C(exchange.GetAccount)
    var detail = JSON.parse(exchange.GetRawJSON())
    if (positions.length > 0) {
        initAccount.Balance += detail['CurrMargin']
    }
    var initNetAsset = detail['CurrMargin'] + detail['Available']
    var initAccountTbl = $.AccountToTable(exchange.GetRawJSON(), "初始资金")

    if (initAccountTbl.rows.length == 0) {
        initAccountTbl.rows = [
            ['Balance', '可用保证金', initAccount.Balance],
            ['FrozenBalance', '冻结资金', initAccount.FrozenBalance]
        ]
    }

    var nowAcccount = initAccount
    var nowAcccountTbl = initAccountTbl

    var symbols = Symbols.replace(/\s+/g, "").split(',')
    var pollers = []
    var prePosUpdate = 0
    var suffix = ""
    var needUpdate = false
    var holdProfit = 0

    function updateAccount(acc) {
        nowAcccount = acc
        nowAcccountTbl = $.AccountToTable(exchange.GetRawJSON(), "当前资金")
        if (nowAcccountTbl.rows.length == 0) {
            nowAcccountTbl.rows = [
                ['Balance', '可用保证金', nowAcccount.Balance],
                ['FrozenBalance', '冻结资金', nowAcccount.FrozenBalance]
            ]
        }
    }

    var q = $.NewTaskQueue(function(task, ret) {
        needUpdate = true
        Log(task.desc, ret ? "成功" : "失败")
        var account = task.e.GetAccount()
        if (account) {
            updateAccount(account)
        }
    })

    _.each(symbols, function(symbol) {
        var pair = symbol.split(':')
        pollers.push(Aberration(q, exchange, pair[0], NPeriod, Ks, Kx, (pair.length == 1 ? AmountOP : parseInt(pair[1]))))
    })

    if (positions.length > 0 && AutoRestore) {
        _.each(pollers, function(poll) {
            poll.restore(positions)
        })
    }
    var isFirst = true
    while (true) {
        var cmd = GetCommand()
        if (cmd) {
            var js = cmd.split(':', 2)[1]
            Log("执行调试代码:", js)
            try {
                eval(js)
            } catch (e) {
                Log("Exception", e)
            }
        }
        tblRuntime.rows = []
        tblMarket.rows = []
        var marketAlive = false
        _.each(pollers, function(poll) {
            if (poll.poll()) {
                marketAlive = true
            }
            tblRuntime.rows.push(poll.getStatus())
            tblMarket.rows.push(poll.getMarket())
        })
        q.poll()
        Sleep(LoopInterval * 1000)
        if ((!exchange.IO("status")) || (!marketAlive)) {
            if (isFirst) {
                LogStatus("正在等待开盘...", _D())
            }
            continue
        }
        isFirst = false
        var now = new Date().getTime()
        if (marketAlive && (now - prePosUpdate > 30000 || needUpdate)) {
            var pos = exchange.GetPosition()
            if (pos) {
                holdProfit = 0
                prePosUpdate = now
                tblPosition.rows = []
                for (var i = 0; i < pos.length; i++) {
                    tblPosition.rows.push([pos[i].ContractType, pos[i].MarginLevel, ((pos[i].Type == PD_LONG || pos[i].Type == PD_LONG_YD) ? '多#ff0000' : '空#0000ff'), pos[i].Price, pos[i].Amount, _N(pos[i].Profit)])
                    holdProfit += pos[i].Profit
                }
                if (pos.length == 0 && needUpdate) {
                    LogProfit(_N(nowAcccount.Balance - initAccount.Balance, 4), nowAcccount)
                }
            }
            needUpdate = false
            if (RCMode) {
                var account = exchange.GetAccount()
                if (account) {
                    updateAccount(account)
                    var detail = JSON.parse(exchange.GetRawJSON())
                    var netAsset = detail['PositionProfit'] + detail['CurrMargin'] + detail['Available']
                    var risk = detail['CurrMargin'] / (detail['CurrMargin'] + detail['Available'] + detail['PositionProfit'])
                    suffix = ", 账户初始净值约: " + _N(initNetAsset, 2) + " , 风控最小净值要求" + MinNetAsset + " , 当前账户净值约: " + _N(netAsset, 2) + ", 盈亏约: " + _N(netAsset - initNetAsset, 3) + " 元, 风险: " + ((risk * 100).toFixed(3)) + "% #ff0000"
                    if (netAsset < MinNetAsset) {
                        Log("风控模块触发, 中止运行并平掉所有仓位, 当前净值约 ", netAsset, ", 要求低于最小净值:", MinNetAsset)
                        if (RCCoverAll) {
                            Log("开始平掉所有仓位")
                            $.NewPositionManager().CoverAll()
                        }
                        throw "中止运行"
                    }
                }
            }
        }
        LogStatus('`' + JSON.stringify([tblRuntime, tblPosition, tblMarket, initAccountTbl, nowAcccountTbl]) + '`\n价格最后更新: ' + _D() + ', 持仓最后更新: ' + _D(prePosUpdate) + '\n当前持仓总盈亏: ' + _N(holdProfit, 3) + suffix)
    }
}

• The test results

  This is because the system is only able to play its power by building a portfolio.

• Summary

  1.The variety and cyclical adaptability of the strategy is excellent, suitable for most trading varieties; 2. the strategy is a large withdrawal and requires strict money management and risk control;