Simple Mean Reversion | Page 16 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

A Simple Mean Reversion Strategy 392 replies

Pairs Trading: Reversion to the Mean 22 replies

Forward test of my new mean reversion strategy 11 replies

Synthetic hedges, cointegration, mean reversion and similar stuff 419 replies

Cointegration, Synthetic hedges, mean reversion in R, Tech Thread 0 replies

Attachments: Simple Mean Reversion
Exit Attachments

Simple Mean Reversion

  • #301
  • Sep 5, 2018 12:58pm Sep 5, 2018 12:58pm
Quoting TheRealDeal
{quote} I have an agreement with my broker about fixed spread, so I test only with fixed. Ofc in order to be able to participate also during rollover when the spread can go as high as 50 pips. MT4 backtester cannot be trusted. I also noticed about it that depending on the strategy and type of test, results can deviate dramatically Suggest you to move also to other software or create your strategy by manual testing and do only forward tests The past doesn’t really matter much
I stopped using MT4 since 2012; and using my own software (java based).

If you are a programmer; it is not difficult to build a backtester that use real ticks. All you have to do is read the file ticks; and loop through them one by one till the end; and spit the results out. You can then improve it bit by bit as you go; till it becomes mature and matches your trading style.

Based on my experience with backtesting on tick data; there is no way GBPCAD will work on this strategy.
  • #302
  • Sep 6, 2018 12:08pm Sep 6, 2018 12:08pm
Quoting alphaomega
Last week I started working on a new improved trading model. The main concept is similar to the original but this time I'm using variable parameters depending on day of the week. I also made new management system which take into account the correlation in a basket of 5 currency pairs. After some extensive backtesting I found that Monday seems to offer the lowest risk profile for opening of new positions. And Thursday is the highest. So the same ( or similar) rules in regards to time cycles are applicable not only for the day but also for the week....
Hi.

Could you try this avarage atr list calculated by me.

for example "eurusd" left side is eur right side is usd. left side eur m30 says 15pips / right side usd says 11 pips. i am in cautious side and i use 11 pips for target and 15 pips for stops if i open trade in m30 tf. (P.S.:and i hope to close in a few hours. 2-3 m30 bars) you could simulate only left or only right.

what will be result.
Attached Image (click to enlarge)
Click to Enlarge Name: Ashampoo_Snap_2018.09.06_19h50m50s_004_.jpg Size: 96 KB
  • #303
  • Sep 11, 2018 7:44am Sep 11, 2018 7:44am
I was on vacation so the week results are a little delayed...

The results from 2018.08.26 to 2018.08.31, nothing spectacular - original method with a little profit, my variant with a little loss.

9990 = my variant of SMR with 4H MA filter, BE stop, no fixed timed 25%/50% exits but "candlestick jumping stop loss" - see above for more details
9993 = original method from post 1 with only BE stop added.
_
EA trades the broker min lots per symbol (0.01 for forex symbols, 0.10 for dax/dow) with some small kind of martingale (0.01 for first 3 trades, 0.02 for the next two trades and then 0.03 lots until MaxTradesPerSymbolAllowed is reached). Both variants open the first trade only when min level of the ATR band of 3 is reached and also have a hard stop if ATR band 6 is reached! The broker used has fixed spreads per symbol and charges no commissions per trade.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 2018-09-03 Simple Mean Reversion EA 9990 kw35.png Size: 10 KB
Click to Enlarge Name: 2018-09-03 Simple Mean Reversion EA 9993 kw35.png Size: 9 KB
  • #304
  • Sep 11, 2018 7:49am Sep 11, 2018 7:49am
The results for the summer month august 2018.

Both methods are nearly identical (both with a loss).
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 2018-09-01 Results August 2018 Simple Mean Reversion EA 9990.png Size: 11 KB
Click to Enlarge Name: 2018-09-01 Results August 2018 Simple Mean Reversion EA 9993.png Size: 9 KB
  • #305
  • Sep 11, 2018 7:54am Sep 11, 2018 7:54am
The first week of september (2018.09.03 - 2018.09.07) ended +-0 for the original method and +50$ for my variant.

9990 = my variant of SMR with 4H MA filter, BE stop, no fixed timed 25%/50% exits but "candlestick jumping stop loss" - see above for more details
9993 = original method from post 1 with only BE stop added.
_
EA trades the broker min lots per symbol (0.01 for forex symbols, 0.10 for dax/dow) with some small kind of martingale (0.01 for first 3 trades, 0.02 for the next two trades and then 0.03 lots until MaxTradesPerSymbolAllowed is reached). Both variants open the first trade only when min level of the ATR band of 3 is reached and also have a hard stop if ATR band 6 is reached! The broker used has fixed spreads per symbol and charges no commissions per trade.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 2018-09-10 Simple Mean Reversion EA 9990 kw36.png Size: 10 KB
Click to Enlarge Name: 2018-09-10 Simple Mean Reversion EA 9993 kw36.png Size: 9 KB
  • #306
  • Sep 11, 2018 8:08am Sep 11, 2018 8:08am
Quoting Jagg
I was on vacation so the week results are a little delayed... Here are the week results from 2018.08.26 to 2018.08.31. Nothing spectacular - original method with a little profit, my variant with a little loss. 9990 = my variant of SMR with 4H MA filter, BE stop, no fixed timed 25%/50% exits but "candlestick jumping stop loss" - see above for more details 9993 = original method from post 1 with only BE stop added. _ EA trades the broker min lots per symbol (0.01 for...
Thanks for providing the ongoing trading results Jagg.

I'm yet to find any solid benefit from using the ATR bands although they do look pretty on the chart.
Pretty or not its the advantage, if any that I'm after and so far its eluding me.

More often than not its simply a case of being fed 'favorable data' by the market that provides the 'nice' outcome.

I've modified the ATR Band formulas but nothing great has been found at this stage.
Including the addition of ADR levels above and below DO price as well as Session Average Range levels above and below DO price to guage SL/or TP.

I've also been testing hourly cycle lines in combination with ATR bands and Fibs to somehow contain the DD and this will continue.

Although as I've always said, open trades 20-30 pips away from the DO line and then close the trade out at the DO line.
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 40 KB

Example,

Stop @ Level 3 or Level 4 (at time of entry)
Entry at Level 2 ATR band
TP @ DO line.

Note that Session Average Range Line is not shown but the actual high and low of completed session is shown (66.7 pips).

Masterrmind.............
Master your Mind then Master your Trades
  • #307
  • Sep 11, 2018 9:11am Sep 11, 2018 9:11am
Maybe most of you don't realize it yet, but I gave you free raw unpolished diamond here!

Attached Image (click to enlarge)
Click to Enlarge Name: 4356546456.png Size: 222 KB


I'm slowly building the strategy behind the scenes and I have confidence that it's going to work very well in the future.

At some point I'm going to present my final version of the complete system along with LIVE account explorer so you can see what is really possible.

Attached Image (click to enlarge)
Click to Enlarge Name: 4354564.png Size: 181 KB


Why this is relevant?

Because the laws of physics govern everything, including the markets!!!
  • #308
  • Sep 11, 2018 11:04am Sep 11, 2018 11:04am
Quoting alphaomega
Maybe most of you don't realize it yet, but I gave you free raw unpolished diamond here! {image} I'm slowly building the strategy behind the scenes and I have confidence that it's going to work very well in the future. At some point I'm going to present my final version of the complete system along with LIVE account explorer so you can see what is really possible. {image} Why this is relevant? Because the laws of physics govern everything, including the markets!!!
Most of FF users need the polished diamond
  • #309
  • Sep 11, 2018 11:17am Sep 11, 2018 11:17am
Quoting alphadude
{quote} Most of FF users need the polished diamond
Sometimes I feel more like they want you to give them the diamond. Oh and please prove it is an authentic one! (and pay for the shipment too)
No greed. No fear. Just maths.
  • #310
  • Sep 11, 2018 11:27am Sep 11, 2018 11:27am
Quoting PipMeUp
{quote} Sometimes I feel more like they want you to give them the diamond. Oh and please prove it is an authentic one! (and pay for the shipment too)
or perhaps a share of alphaomega profit directly to my MT4 account with comments on transaction.

Inserted Video
  • #311
  • Sep 11, 2018 12:00pm Sep 11, 2018 12:00pm
Quoting alphaomega
Maybe most of you don't realize it yet, but I gave you free raw unpolished diamond here!
Give them the right {timing function},
and here you go, the diamond gets... polished!
Maadik Hugiis. IQ 69.
  • #312
  • Sep 11, 2018 1:21pm Sep 11, 2018 1:21pm
Quoting alphadude
https://www.youtube.com/watch?v=QRCLWl1WRV0
Buy when the candle opens on its low.... I'm still laughing...
No greed. No fear. Just maths.
  • #313
  • Sep 12, 2018 12:48am Sep 12, 2018 12:48am
Quoting Masterrmind
{quote} One reason why I code my own and pretend I know nothing ! If only they knew that a simple thanks means so much more than they could possibly imagine. ps. if they coded it themselves they would soon understand that its more than a 30 second job. Masterrmind..............
this is very deep.
Appreciating someone's work is very significant.
Noting is wrong saying thanks to someone.
Profit growth tells you the trader ability.
  • #314
  • Sep 14, 2018 9:22pm Sep 14, 2018 9:22pm
Quoting Jagg
{quote} You mean something like this? {file} {image}

Hi Jagg, thank you very much for sharing your indicator. It is great to looking back the historical data and find it very useful to filtering those PA during Tokyo and early Europe market sessions. On the other hand, the indicator will repaint slightly, with a variance of 1 to 2 pips, over the day until the day close then no more repaint occur for that day. I guess the reason is the use of ATR which has included the TR of the current day (as day high/low will change over time). Is it possible to use ATR of previous day to keep the bands constant over the day?
  • #315
  • Sep 16, 2018 12:52am Sep 16, 2018 12:52am
Quoting TheRealDeal
{quote} Lately I am too paranoid about any commercial or free software. I am writing my own now
This why I trade manually always. No automation of any sort, at any time. Do I miss trade levels, sure do. Price I pay for the life I choose.
If you are good at something, never do it for free--Joker
  • #316
  • Sep 16, 2018 1:02am Sep 16, 2018 1:02am
The whole concept of mean reversion is straightforward and easy to understand.
I don't trade full mean reversion strategies live yet, but I intend to add a pure mean reversion strategy to my current portfolio.
I have attempted to create a fully automated EA with OP's original rules - with limit orders and market orders. But a intraday close all rule just wasn't giving me the ideal profit scenarios. My next step is to extend the intraday cycle using the original entry rules, but with different exit options.
  • #317
  • Sep 16, 2018 1:03am Sep 16, 2018 1:03am
Quoting jmn5611
{quote} This why I trade manually always. No automation of any sort, at any time. Do I miss trade levels, sure do. Price I pay for the life I choose.
Wouldn't you want to pass something tangible to your kids as a foundation for their future? That was my only intent to move from manual to automation plus travel the world care free and have someone else monitor my automation :-)
Staying in my lane...
  • #318
  • Sep 16, 2018 1:22am Sep 16, 2018 1:22am
Quoting VEEFX
{quote} Wouldn't you want to pass something tangible to your kids as a foundation for their future? That was my only intent to move from manual to automation plus travel the world care free and have someone else monitor my automation :-)
That is very interesting that you would say that, because that is one of my major drivers.

1. I told my kids that I started late. A 20 year old with my information is almost promised to be a millionaire.
2. I distrust automation because platforms change with OS. In short, if you have a system that will bleed them dry slowly but surely you will be met with the same fate that happened to the MIT kids who cracked the blackjack table. You would never want to automate that.
3. Because of the WWW, I can still travel at will, I only really need to be online once a day at the FX reset. I just do not trust automation. Too many things can go wrong. That said, automation is one way to defeat biases.
If you are good at something, never do it for free--Joker
  • #319
  • Sep 16, 2018 4:13am Sep 16, 2018 4:13am
Quoting jmn5611
{quote} That is very interesting that you would say that, because that is one of my major drivers. 1. I told my kids that I started late. A 20 year old with my information is almost promised to be a millionaire. 2. I distrust automation because platforms change with OS. In short, if you have a system that will bleed them dry slowly but surely you will be met with the same fate that happened to the MIT kids who cracked the blackjack table. You would never want to automate that. 3. Because of the WWW, I can still travel at will, I only really need...
Automation can be used in a smarter way than trading. For example if you are looking for an edge or to add a tool to your manual trading. Lets say if you trade Daily chat and you want to use a reversal candle pattern, but you dont know which is better : a pinbar or engulfing candle; inside or outside bar; etc. So you write a simple program that will give you the results over thousands of examples in minutes. You will spend weeks to figure out that manually.

When you create, upgrade or just improve your strategy, you need to take out every feature and test it independently how it performs. For that automation is useful.

TRD
  • #320
  • Sep 16, 2018 4:40am Sep 16, 2018 4:40am
Real proven edge + Automation + OPM(Other People's Money) = BIG SUCCESS!

You still have doubts about this??
Inserted Video


But still, probably the best solution for the average trader is the hybrid system. Automation + discretionary.

I'm also big believer in pure old school manual trading because this style has its own advantages. However, the problem with manual trading is that it creates a lot of stress and emotions. And you get tired very quickly, especially with intraday trading and scalping. And when you get tired - you make mistakes!

The robot never gets tired, and never makes mistakes.
0 traders viewing now