Hi SFF,

I would say it's imperative to having a successful strategy..

The last thing you want is for a trade to become negative from the volatility of the forex markets so narrowing this down will go some way to helping your result. There are many mm methods are we know from the manual.

I guess the possibilities are endless, but having lost money from trading in the past due to emotions I find that having moving into algorithmic trading is maybe the answer to actually making money rather than loosing it..

That's why I have been observing the markets for the past 12 months without actually trading and looking into BOTS because I knew if i traded again my emotions would most likely end to losses in the end.

Now we have Zorro so the average man now take it upon himself to work on a strategy that will profit from the markets and there are no emotions involved.

Kind regards.

Hi,

>>How to do that is explained in the Zorro tutorial.

I am looking for a tut for Spectrum.
Where can I find it?

Are spectral analysis effective for a lower TF like 1 min?

There is no tutorial, only the function description under "Filters" and the example script spectrum.c.

With low time frames the spectrum becomes flat and loses its peaks. Spectral analysis works best with time frames in the hour or day range.

You said "How to do that is explained in the Zorro tutorial." in this page.
http://zorro-trader.com/manual/en/strategy.htm

That refers to workshop 5 that uses the dominant cycle for trading. The dominant cycle is the main peak of the spectrum.

Hi,

I have been able to see some manual traders making over 10,000% in short priod but
not automated traders?

Why this happens? I am sure I can make such % with robots too.

You need no robot for this. For making over 10,000%, just go to Las Vegas and bet twice on Roulette. For this you need not as much luck as that manual trader.

I'm working my way thru the Jackal and Tomansini book and I understand that the strategy is not profitable.
After they present the strategy as you have written above, the next step in the book is to "optimize" the "profitabe" strategy by adding a timefilter.
in the book's example they study a 4 hour trading time window shifting the starting time 30 minutes thoughout the day.
I was going to try var starttime = optimize() but some of the values can't be zero. When you use var endtime = optimize() the clock isn't linear so that won't work.
I found the hour() time variable and can make that work for a specific time period, but I'm having trouble trying to shift the starting times as they do. in the book. The only idea I have is an array with each start time and end time defined and then go thru the array 1 time period at a time. Any other ideas?

Thanks

P

Multiple time filtering with an array would likely lead to an overfitted and unusable strategy.

Optimizing a start time and a time window is better, but I would use time filters with extreme caution - they can let any losing system appear profitable in the backtest. Anyway, for optimizing something that can be 0 or negative, just subtract a constant value like this:

int start_hour = optimize(1,1,24,1)-1;

Following on from the book, how does one then optimize the paramaters 'Fast' and 'Slow'?

My attempts fail big time. After setting the Paramaters flag, I attempted to insert the optimize function for the Fast moving average like so.

vars Price = series(priceClose());
vars Fast = ptimize((1,1,20,1)series(SMA(Price,3));
vars Slow = series(SMA(Price,30));

The error message is as follows.

Luxor compiling.............
Error in 'line 23: syntax error
< vars Fast = optimize((1,1,20,1)series(SMA(Price,3));
>.