Discrepancy Between Advanced Chart and Back Testing

[mebstyne]

Recently I had a chance to evaluate eSignal. I'm debating between eSignal and TradeStation. I'm leaning towards eSignal, but still clearly on the fence.

During my evaluation, I had a chance to play around with EFS and backtest some strategies. The EFS script itself uses a visual indicator (a dot over/under a bar) to indicate when the strategy goes long/short.

When I compare the list of trades entered in the strategy analyzer with the visual rendition of the strategy applied to an advanced chart window, the exact trades don't match. For example, the strategy analyzer might read a trade is entered at 11/28/04 at 1:40PM and there are NO OTHER TRADES that day according to the analyzer's report. Now, if I look at the chart, I might see two or more trades entered on the 28th, neither of which are particularly close to the 1:40PM.

Both are using the same Time Template.

I looked at repeatedly over a few days. This descrepency between the analyzer and the chart seemed to occur on all the strategies that I examined closely.

I'd much appreciate it if anyone can share insights into what the issue might be? Surely, I'm not the only one who has had to face this or something like this early on in their days of programming EFS?

Thanks,

mebstyne

[ACM]

mebstyne
Can you post any one of the efs(s) used for your tests and provide the security, interval and Time Template settings so as to try and replicate the issue?
Alex

[mebstyne]

still looking

It's been a while since the last post, I thought I'd provide a quick update.

Although I am still seeing unexplainable discrepancies, I have been unable to boil them down into a simple example. In the process of reducing the noise to isolate the issue, the seems to go away.

In the process, I ran into an interesting parsing error. I will post that in a separate thread.