## Grid build with FxGen

By forexnes 2388 days ago Comments (6)

Hello everyone,

This is the beginning of a simple idea for EA grid that tracks iMA20 < iMA30 is short and opens whenever the price EURUSD Bid=1.3300 in the range of pairs = 1.3300 -1.3070. Whit profit 10pips. If someone is a programmer and is willing to further develop the idea, please write :)

• jaguar1637 2388 days ago

Ok I will quickly correct this EA and post it, it will take 5 mn to correct it

• JohnLast 2388 days ago

Thank you jaguar for the help.

forexnes as you are Bulgarian you can refer to the vgc material about the mathematical model of the grids written by vgc.

I realy need that it will be good to make a proper English version of those posts.

The problem withthe forums is that very often some very intersiting things are lost in the jungle of rubbish.

• forexnes 2388 days ago

Thanks for the link and support JohnLast and jaguar1637 :)

• JohnLast 2388 days ago

I will resume the core idea of VGC about the grids.

Wat really matters is to set the grid, the cell size of the grid. According to vgc

The optimal grid cell size is optimal to set equal to the hourly volatility.

-If the grid size is too small you will get hurt by the spread.

-If the grid size is too large does not lead also to more profitable results as you miss number of profitable trades.

There are two axioms:

1. The relationship between the volatility of different frames is equal to the square root between them.

So if we have an hour volatility of 20 pips the daily voltility would be: 100 = 20*sqrt(24)

2. During a day there would be at least 12 movements with the size of the hourly volatility.

An example of a classical Grid system with 20 pips cell size equal to the daily volatility.

We expect to trade for a year.

The expected volatility is:

-average hourly volatility 20 pips

-average daily volatility 100 pips

-average annual volatility 1600 pips

The optimistic costs per trade are 4 pips. For practical simplification each pips is equal to 1 dollar.

Numbers:

-We have 260 working days per year., and we expect to have 3120 positions.

-The opened loosing trades at the end of the year (31.012 ) would be -\$32 500.

We can model it as the half of the aderse movement of \$65 000 equal to the maximum volatility of 1600 pips and during this time we would 80 orders contrarian to the unidirectionnal movement.

-The close trades would accumulate + \$ 65 000 as a result.

- Costs for the spread   \$12 500

As total the gain is \$16 500.

We would need optimistically 2* \$65 000 = \$130 000 as a balance in the beginning of the year.

Conclusions:

We can expect profitability at the end ofthe year at average 12 %.

Any attempt to improve the expected profitability would be severy punished by the market.

According to vgc, the profitability in practice is 10 % and does not corresponds to the amount of the risk involved with that system.

• JohnLast 2387 days ago

I feel some of those developments merit a separate blog post.

However I will just share an idea, it is not mine it is from a porfessional guy, who just dropped the idea by the way.

I will cite one opinion of Mr. SB.

My use of the term coin flip is that I don't know what price is going to do but I have patterns that allow me to have an edge anyway.

My edge is mostly mathematical in that I don't have to get it right, I only have to avoid getting it completely wrong. If price moves as predicted (lets call that heads) I win in the obvious way. If price goes against me (tails), because Icalculate levels and not specific prices the algorithms will normally get me out at a small profit or breakeven. If it loses, it will be a small loss.

Think about Hurst exponents; many do not know the original use was in building dams. It wasn't designed to predictwhere something will go as it was to predict where it is unlikely to go (water levels).

Should anyone feel inclined to give what I do a different name, a wikipedia reference or even their opinion on how I can it even better please feel free.

In the end it pays the bills and has for a long time.

So here it is.

The Studies involving the Hurst exponent were originally developed in hydrology for the practical matter of determining optimum dam sizing for the Nile river's volatile rain and drought conditions that had been observed over a long period of time.

So applied to the price the Husrt exponent is not telling us where the price will go, but where the prices are not likely to go.

So there are two possible implementations I have in mind:

1. So this can be applied to a grid expert, considering the maximum adverse excursion as river dam limit.

2. Te second idea is to use the local Hurst exponent as a way to achieve an adaptive trailing stop loss level (support and resistence levels).

• forexnes 2387 days ago

Thanks for this information:)