RSI Mean Reversion Trading Strategy

Author: ChaoZhang, Date: 2023-09-12 14:37:28
Tags:

This strategy is based on the mean reversion characteristics of the RSI indicator. Overbought and oversold RSI tends to revert back, creating trading opportunities. The strategy identifies overbought/oversold states using RSI to establish long/short positions in a systematic way.

Strategy Logic:

  1. Calculate RSI value and set overbought and oversold thresholds, typically 60 and 30.

  2. When RSI crosses down the overbought line, go short.

  3. When RSI crosses up the oversold line, go long.

  4. Long stop loss is entry price * (1 - stop loss %). Short stop loss is entry price * (1 + stop loss %).

  5. If price hits the stop loss, exit the position.

Advantages:

  1. Captures mean reversion opportunities during trend pullbacks using RSI.

  2. Breakout trading allows timely entry at trend reversals.

  3. Stop loss controls single trade loss amount.

Risks:

  1. RSI tends to give false signals. Confirm with other indicators.

  2. Stop loss being too tight causes excessive stops. Consider widening range.

  3. Poor timing of entries may lead to oversized positions.

In summary, the RSI mean reversion strategy trades RSI overextensions. It follows the trend with controlled loss on single trades. But RSI reliability is low. Investors should use it prudently with other confirming indicators, optimized stops, and expect modest long-term returns.


/*backtest
start: 2022-09-05 00:00:00
end: 2023-09-11 00:00:00
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/
// © relevantLeader16058

//@version=4
strategy(shorttitle='RSI Bot Strategy',title='Quadency Mean Reversion Bot Strategy', overlay=true, initial_capital = 100, process_orders_on_close=true, default_qty_type = strategy.percent_of_equity, default_qty_value = 100, commission_type=strategy.commission.percent, commission_value=0.08)

//Backtest dates
start = input(defval = timestamp("08 Mar 2021 00:00 -0600"), title = "Start Time", type = input.time)
finish = input(defval = timestamp("9 Mar 2021 23:59 -0600"), title = "Start Time", type = input.time)
window()  => true       // create function "within window of time"

// Complete Control over RSI inputs and source price calculations
lengthRSI = input(14, minval=1)
source = input(title="Source", type=input.source, defval=close)
strat = input(title="Strategy", defval="Long/Short", options=["Long Only", "Long/Short", "Short Only"])
strat_val = strat == "Long Only" ? 1 : strat == "Long/Short" ? 0 : -1
stoploss = input(5.00, "Stop Loss %")
oversold= input(30)
overbought= input(60)

// Standard RSI Calculation
RSI = rsi(close, lengthRSI)
stLossLong=(1-(stoploss*.01))
stLossShort=(1+(stoploss*.01))

//Long and Short Strategy Logic
GoLong = crossunder(RSI, oversold) and window()
GoShort = crossover(RSI, overbought) and window()

// Strategy Entry and Exit
if (GoLong)
    if strat_val > -1
        strategy.entry("LONG", strategy.long)
    if strat_val < 1
        strategy.close("SHORT")
    

if (GoShort)
    if strat_val > -1
        strategy.close("LONG")
    if strat_val < 1
        strategy.entry("SHORT", strategy.short)


LongStopLoss = barssince(GoLong)<barssince(GoShort) and crossunder(low, valuewhen(GoLong, close, 0)*stLossLong)

ShortStopLoss = barssince(GoLong)>barssince(GoShort) and crossover(high, valuewhen(GoShort, close, 0)*stLossShort)

if (ShortStopLoss)
    strategy.close("SHORT")
    
if (LongStopLoss)
    strategy.close("LONG")






More