I have never used the RefreshRates() before. The for loops was doing that. at least attempting to do that...

I ran backtests from '04 and in July of '04 the errors still occurred. I will look closer at the code and see if I can find anything.

Don

Version 2 seems to work the best of all the versions thus far. It lacks the Trend that will make the Margin safer, but the execution is best. I have added tidbits to update the code to include some of the updates that Neo has posted...

Neo -

I would like to implement a Trend to this version. I have been using a LWMA to achieve this. However, the close routines go all to hell with the introduction of the MA... Maybe you might have some ideas to incorporate it into the EA successfully.

Don

Attached Files

PipMakerv2a.mq4 (19.3 KB, 348 views)

I'll take a look....

What I have noticed with the PM2a is that the opposites work only when the price is at the very extremes. Higher than the HighestBuy or lower than the LowestSell... when the positive sell are above the MidPoint, the losing sell is NOT closing. same for the low buys not closing out the high buy... who would of thought simply adding a iMA would screw it up so badly...

Don

Good morning Don,

Firstly, sorry for my bad english, i don't speak good english.

Secondly, thanks to share your work with us.

So i'm very interest about your EA, so i want to test it in live on a demo account (3000$ Lot: 0.1), just want to run it while 3 at 5 months to see the results, so i want to know what is your best timeframe to run it.

Thanks

Ok...found part of the problem, at least! The MinWait computation in the "Close in profit" functions is not fully supported in Strategy Test mode so replace the functions with these:

void CloseBuysinProfit()
{
double BuyProfit, LastBuyTime;

for (int i = OrdersTotal() - 1; i >= 0; i--)
{
if (OrderSelect(i, SELECT_BY_POS, MODE_TRADES))
{
if (OrderSymbol() == Symbol() && OrderMagicNumber() == MagicNumber && OrderType() == OP_BUY)
{
BuyProfit = OrderProfit() + OrderSwap() + OrderCommission();
if (OrderOpenTime() > LastBuyTime) LastBuyTime = OrderOpenTime();
if ((IsTesting() || TimeCurrent() - LastBuyTime >= MinTime) && BuyProfit > 0) OrderClose( OrderTicket(), OrderLots(), MarketInfo(OrderSymbol(), MODE_BID), 5, Red );
}
}
}
return;
}

//+-----------------------------------------------+
//| Closes all Sells in Profit
//+-----------------------------------------------+
void CloseSellsinProfit()
{
double SellProfit, LastSellTime;

for (int i = OrdersTotal() - 1; i >= 0; i--)
{
if (OrderSelect(i, SELECT_BY_POS, MODE_TRADES))
{
if (OrderSymbol() == Symbol() && OrderMagicNumber() == MagicNumber && OrderType() == OP_SELL)
{
SellProfit = OrderProfit() + OrderSwap() + OrderCommission();
if (OrderOpenTime() > LastSellTime) LastSellTime = OrderOpenTime();
if ((IsTesting() || TimeCurrent() - LastSellTime >= MinTime) && SellProfit > 0) OrderClose(OrderTicket(), OrderLots(), MarketInfo(OrderSymbol(), MODE_ASK), 5, Red);
}
}
}
return;
}


Neo

I have replace those subroutines. I will try to add in the LWMA now and see if the closes work better...

Don

The addition of the LWMA still screws up the closes. Frustrating.
All I did was create an LWMA variable.


extern int MAPeriod = 1200;

double Trend = iMA(NULL, 0, MAPeriod, 0, MODE_LWMA, MODE_CLOSE, 0);

And then call it up in the Buy and Sell trade criteria.

if (Ask < LowestBuy-(Spacing*Point) || Ask > HighestBuy + (TrendSpacing * Point) && Ask > Trend)

if (Bid > HighestSell + (Spacing * Point) || Bid < LowestSell - (TrendSpacing * Point) && Bid < Trend)


Don

I use 1M time frame as there isn't any indicators involved.

I am trying to incorporate a LWMA and still will use a 1M tf.


Don

Hi Don !!
I an shure that is a very good idea. I am backtesting in 30 min TF and getting interesting results (profit factor 3+ but a little high drawdown). I will forward test in that TF and post my results latter.

Regards,