From what I recall its an automated process, so perhaps the email with your serial number is landing in the spam folder. You could just try a different email provider altogether. Its definitely worth checking out the trial. Mine expired yesterday and I considering buying it at some point if the scripts it made wind up being useful .
Is this fully optimized for the entire date range? If so, this will likely immediately fail when going live. Its best to leave a good amount of an "Out Of Sample" period and see if it works on that. For example, you having left out 2014 altogether and tested the strategy on that to simulate as if you started trading it one year ago. This will show you what would had happened.
Another great thing is walk forward optimization. This can be done in NinjaTrader directly.. Basically it will let you know if the strategy will hold up over time after you periodically reoptimize every X amount of months. Every strategy is different, one of the ones Im working on at the moment passed reoptimizing every 56 days with the optimization done over the prior 4 months. I ran this for the past 3 years and it was generally mostly profitable.
Last thing to note, be careful with this Ninjatrader flaw. When you use the ninja trader walk forward analysis, make sure to select time intervals of seven days and only start the date range on a Sunday. The reason why I suggest this is because if you start midweek and you don't use intervals of seven days then the walk forward optimization will overlap trades and give you results that seem much better than they really are.
In other words trades that should've ended in period One continue on to period two and vice versa. When you review the trades you will notice that the same trades exist at the end and beginning of optimization phases. Something that normally would not happen in real life obviously.
So by choosing dates with intervals of seven forces the walk for analysis to close positions on Friday and never overlap them as there is no trading on a Saturday. The other option is to use a session template that does not start the prior day but at midnight, but this method I found to be inaccurate when compared to real life as we will never end positions at 11:59 PM.
Ninja trader said that this bug is due to the session template of most futures starts the day before as opposed to within the same day and they are the ones that suggested to create a special session template that goes from midnight to 1159pm this way the forward analysis will not overlap trades. This bug is most noticeable if you try to walk forward analysis one day at a time. As an example is, you try to do a daily walk forward analysis with optimization from the prior five days.
Sorry to take so long to reply. The strategy was developed on 6B and tested on CL. The program sort of optimizes when the strategy is evolved. But I can get better results optimizing in NinjaTrader on 6B and CL. The walkforward give about half the optimization with slippage set at 2. Still working with the program as it is not at all obvious how to use the various settings.
That article is really useful. I especially like the part about removing the exit strategy as a test and using a small profit target and stop (Should both be equal, perhaps 2 ticks each).. This will give you a great indication of how your entries are doing by looking at the "% Profitable" column of the backtest. Of course one can't look at profit factor and other details in the report since this is specifically for judging the entry accuracy but never the less brilliant.
The following 2 users say Thank You to danayster for this post: