Search found 230 matches

by Tim Arnold
Wed Nov 12, 2008 7:02 pm
Forum: Testing Software
Topic: Trading Blox purchase?
Replies: 11
Views: 13126

To receive the trial version information and download link, send an email to freetrial@tradingblox.com.
by Tim Arnold
Fri Aug 15, 2008 8:14 pm
Forum: Data Providers and other non testing software
Topic: CSI Default screen?
Replies: 11
Views: 9528

This sounds like a problem we sometimes have with Vista and the Program Files directory. Did you install UA in the Program Files directory and are you using Vista?
by Tim Arnold
Mon Aug 04, 2008 8:32 pm
Forum: Data Providers and other non testing software
Topic: Difference Forex EOD Metastock and UA
Replies: 1
Views: 3171

They might use a different closing time, or source of data. Check with CSI as they can probably tell you right off.
by Tim Arnold
Wed Jul 09, 2008 8:03 am
Forum: Testing and Simulation
Topic: Calculate Max DD in dollars
Replies: 3
Views: 3751

Loop over the equity curve and keep track of the high, and the following low, and compute the dollars.
by Tim Arnold
Sat Jul 05, 2008 6:02 pm
Forum: Testing Software
Topic: Sorting Custom Arrays
Replies: 6
Views: 8357

I have added the ability to sort custom arrays. This feature will be available in the next beta release for testing. Please post examples of usage here so we can be sure the architecture will meet your needs.
by Tim Arnold
Fri Jun 27, 2008 8:46 am
Forum: Testing and Simulation
Topic: Trading Blox questions
Replies: 5
Views: 4678

Sure, give me a call anytime. The number is on the website.
by Tim Arnold
Wed Jun 25, 2008 6:27 pm
Forum: Testing and Simulation
Topic: Trading Blox questions
Replies: 5
Views: 4678

Trading Blox does not automatically create the four hour bars from the one minute bars. But you could check the time and see if it is on a four hour interval and then generate the signals. Or you could create the four hour bars OHLC through scripting using the Series Variables. Depends on how comple...
by Tim Arnold
Sat Jun 14, 2008 4:48 pm
Forum: Testing and Simulation
Topic: Update CSI data without overwriting files?
Replies: 3
Views: 3677

Sounds like a question for CSI.

My answer would be "can't be done"
by Tim Arnold
Mon Jun 09, 2008 10:43 am
Forum: Testing and Simulation
Topic: Turtle System results
Replies: 6
Views: 10226

If you use 1 unit max rather than 4, it will be ok -- but try the Donchian system as that is a simliar system but with much better results due to the macd filter.
by Tim Arnold
Thu May 29, 2008 12:16 pm
Forum: Testing and Simulation
Topic: Optimization paradox
Replies: 1
Views: 2873

No further articles were written on the topic, or any other topic.
by Tim Arnold
Wed May 21, 2008 6:05 pm
Forum: Testing and Simulation
Topic: Recording trades in TB
Replies: 9
Views: 8380

From this, I gather that you either need to enter all of your own trades or use the TB simulation without any adjustment - mixing the two isn't accurate at this time. Trading Blox allows you to enter current Positions, not a historical list of Trades. So in Trading Blox it is appropriate and accura...
by Tim Arnold
Tue May 13, 2008 4:09 pm
Forum: Testing and Simulation
Topic: Common sense defied...
Replies: 11
Views: 7622

When I hear "can't think of any reason" my reaction is "not statistically significant."

Then again, there could be a logical reason...
by Tim Arnold
Sat Apr 19, 2008 6:00 pm
Forum: Testing and Simulation
Topic: Backadjusted point or ratio (%)
Replies: 9
Views: 8299

The sample data is back adjusted as described above. We do not recommend using ratio adjusted data for back adjusting futures as you will get incorrect P&L computations.
by Tim Arnold
Mon Apr 14, 2008 12:31 pm
Forum: Testing and Simulation
Topic: MAR Ratio at +Infinity
Replies: 2
Views: 3062

Looks like you have a drawdown of zero, which results in a MAR of infinity. Your CAGR is probably related more to interest earned than to actual trades.
by Tim Arnold
Mon Apr 14, 2008 9:21 am
Forum: Testing and Simulation
Topic: Correlation vs. Correlationlog
Replies: 7
Views: 5803

The reason to use thing A vs. thing B is because thing A works better for your purposes. So I would recommend testing your correlation strategy using both and see which is a better fit for your needs. That said, I find people often use the log functions for correlation and for standard deviation whe...
by Tim Arnold
Wed Apr 09, 2008 12:37 pm
Forum: Forex
Topic: Create my own forex data?
Replies: 6
Views: 10903

Here's some data of each cross currency, the actual JPYCAD and a calculated JPYCAD. Looks close, but not exact. Room for arbitrage I suppose. Date USDJPY USDCAD JPYCAD Calculated 20080401 101.96 1.021500 0.010017 0.010019 20080402 102.36 1.014100 0.009900 0.009907 20080403 102.32 1.004700 0.009816 0...
by Tim Arnold
Fri Apr 04, 2008 5:01 am
Forum: Testing Software
Topic: Monthly Bars
Replies: 1
Views: 4618

In the current version 2.2 Trading Blox does not compute the monthly bars from the daily at this time. Customer tend to trap the month change in the After Instrument Day script and keep track of this in a few series variables. In version 2.3 Trading Blox supports Daily (from intraday), Weekly, and M...
by Tim Arnold
Thu Mar 20, 2008 6:17 am
Forum: Trader Psychology
Topic: This is why trendfollowing is hard
Replies: 19
Views: 25823

Why not take some money of the table and play with a smaller trading capital. IF that was part of your system design and you backtested this scenario and you are comfortable with the results, then this is fine. But if you did not backtest taking money off the table, then unfortunately that option i...
by Tim Arnold
Thu Mar 13, 2008 6:46 am
Forum: Testing and Simulation
Topic: Toronto area Reference for TradingBlox please
Replies: 5
Views: 4418

We actually have a number of customers in the Toronto and Ontario areas, not sure why they are being so shy. Sorry about that. I find many traders tend to be quite secretive.
by Tim Arnold
Fri Feb 01, 2008 5:59 am
Forum: Testing and Simulation
Topic: Anybody using Quad Core Machine for testing
Replies: 28
Views: 26120

Would be interesting to know under a 32 bit - 3GB or x64 - 10GB framework which are maximum database sizes in term of number of stocks and number of years. Exact numbers depend on the type of system, how many indicators, how many trades, etc. But I run 6000 stocks for 10 years with just 2GB of memo...