Performance report trades tab options

[MrSmartGuy]

Does anyone know of any report content options for the "Trades" tab on the performance report? I'm seeing two data variants in the MAE and MFE columns. Sometimes the excursion data appears as the true intra-trade max excursion, as it should, and then other times it shows the simple trade profit/loss value. But, yet, I have not changed anything. So, maybe there is a switch or configuration option somewhere that I need to learn about?

[NinjaTrader_Kyle]

Hello,

Thank you for your post.

There are no options related to MFE or MAE in this menu which can be toggled. Can you provide me with a more specific example of the issue?

[MrSmartGuy]

Hello Kyle,

I have attached a screen shot showing the anomaly. See trade 31 in the Trades Report, corresponding to the last trade on the daily chart. The MAE shows zero in the report, but should be $3,675. The MFE shows $545, which is the trade profit, rather than the higher max excursion level.

Thanks for looking into this.

[sledge]

I have a market replay run open in NT7..

My strategy performance real time trades MAE/MFE is correct, and matches Control Center->Account Performance->Trades tab. You can do a generate in there.

So does that match for you?

Also - if you goto Graphs in either area, and then MAE/MFE - is it the same as the data columns (ie, incorrect)?

Where did you run this?

And which version are you using? 7.0.1000.xx? 32 or 64 bit?

And is this an initial run from starting NT7 or is this like the 2nd-10th run and results?

[MrSmartGuy]

Thank you for your interest, Sledge.

The control center report is unpopulated, both before and after generation.

The MAE/MFE graphs do match the data in the columns, so a common source is apparent.

I ran this on a reasonably modern PC (2.8 GHz clock, 4 GB RAM, 64-bit OS) with NT 7.0.1000.29.

I have been working on this code for a couple of months now, with few problems of this sort. But I have been focusing on other performance parameters in the reports, so I can't say exactly when the MAE/MFE fields went awry.

One further clue. A month ago I was having trouble getting the Summary Tab drawdown parameter to work, which is when I first looked at the detailed MAE/MFE values and saw that they were not correct. The home office suggested corrupt data as the culprit, most likely due to a power interruption. So, I reloaded the data base, and it fixed the MAE/MFE values. I assumed that the reload had done the job, and I regret not occasionally rechecking them since then, as I now see them wrong, but I don't know when things went wrong.

Given the massive number of ways that NT can be configured and used, I began to wonder if I was just missing a configuration parameter somewhere, and so I put up this post.

[MrSmartGuy]

For NinjaTrader_Kyle:

Hi Kyle,

I've experimented a bit more and discovered that the MAE/MFE values are correct for the first time that I run NT after clearing the data base (deleting the NinjaTrader.sdf file prior to starting NT.) Once I exit NT and then restart, however, the MAE/MFE values revert to the wrong values.

I have attached a Word document showing examples of each case.

Thanks for looking into this.

[NinjaTrader_Brandon]

Hello MrSmartGuy,

Thanks for your reply.

I have attempted to reproduce this with the SampleMACrossove however I could not.
If you complete the same test using the SampleMACrossover can you reproduce this?

[MrSmartGuy]

Hello Brandon,

Attached is a Word document showing test results for the experiment that you suggested. The first screenshot shows the performance report Trades tab for three test cases using the first NT load after the data base reset. The cases are my original strategy, replicating what I reported in the 9-14 document, and two additional reports for AAPL and SPY using the SampleMACrossover strategy, configured for 3000 days with a 50/200 cross.

I used SPY to see if maybe the anomaly is SPY related, which is what I am using in my custom strategy. I also use AAPL in a separate chart, as it has never been used before. In all three cases, the MAE/MFE values are good in the first test and in error in the second test. The errant values all reflect the simple PnL for each trade.

I appreciate your looking into this. Please advise if further testing may help. This is quite the mystery.

Brad

[NinjaTrader_Brandon]

Hello MrSmartGuy,

Thanks for your reply.

I have been able to replicate this and we will look further into this.
I would expect that this will be looked into for NinjaTrader 8.

[shuglu]

NT is well aware of the fact that both MAE and MFE are routinely incorrect.

A search on the topic will show that this question often arises and and likely not dealt with since inception as not a priority for most.

I access the sdf database itself for my calculations. When running a one tick chart for testing supposes it clearly shows that NT does not include maximum and minimum excursion on one contract between entry and exit on a discretionary basis.

I can only assume that a decision has been made not to act on this shortcoming.

[NinjaTrader_Brandon]

Hello shuglu,

Thanks for your reply.

Most reported items at this time are only being looked into with NinjaTrader 8. NinjaTrader 8 is developments main focus.

[MrSmartGuy]

Brandon,

So, if I hear you right, bugs like this one are a secondary priority at this time, with NT8 original development activities as the first priority. Have I got that right?

You noted earlier that you expected this bug would be looked into for NT8. Not to parse words to too fine a level of detail, but that is not the same as saying that it will be looked at. Bottom line, then, is what is your best estimate of the probability that this bug will be fixed in NT8?

Thanks.

Brad

[NinjaTrader_Brandon]

Hello Brad,

Thanks for your reply.

Development are currently looking into these statistics with NinjaTrader 8. I could not say when this will be resolved in NinjaTrader 8.