Search results

  1. N

    EFX Group

    They're quite clear about this when you sign up for an account: "The transactions you are entering into with mb trading futures are not traded on an exchange. Therefore, under the u.s. bankruptcy code, your funds may not receive the same protections as funds used to margin or guarantee...
  2. N

    EFX Group

    I have a question on EFX. I've just started using this platform, so please forgive my ignorance. I'm currently short a position and want to add a stop-loss. When I enter a buy order with a stop, the order is refused because of insufficient buying power. Why is that? If the order is executed it...
  3. N

    Statistics Question

    The classical textbook is Bradley Efron, Rob J. Tibshirani. An Introduction to the Bootstrap. Wikipedia also has some useful information
  4. N

    Random Trade Generator

    you're welcome. good trading to you!
  5. N

    Statistics Question

    sorry that I overlooked your question. It is called a nonparametric bootstrap.
  6. N

    Random Trade Generator

    I posted an example spreadsheet in another thread that you might find useful: http://www.elitetrader.com/vb/showthread.php?s=&postid=1246313#post1246313 I generated the replications using Excel's Data Table functionality. (It's a very cool and powerful feature useful in lots of situations...
  7. N

    Random Trade Generator

    ok, let me try to explain the confidence interval very briefly. Suppose you calculate a statistic (e.g. expectancy) over all your trades up to today, and find it's 3.4753% Now you make one additional trade tomorrow. Guess what? Including the new trade it changed to 3.4780%. So, your statistic...
  8. N

    Random Trade Generator

    no, it's not just rearranging (in that case statistics like the profit factor or expectancy wouldn't change with each new run, also the endpoints of all your curves would be equal). Remember, if you draw 100 random numbers between 1 and 100, it's likely that some are equal. So, each run...
  9. N

    Random Trade Generator

    One possible approach is resampling (bootstrapping): - Suppose you have 100 trades, each with return R. - You draw a random number N between 1 and 100 and take the Nth return. Repeat this 100x. - You now have a new set of trades drawn from the distribution of your original set. If you plot...
  10. N

    how would you sell a working system, if you had a better one

    You can contact the owner, Matthew Klein, directly through the site and discuss your situation with him.
  11. N

    how would you sell a working system, if you had a better one

    They only show *closed* trades to people that are not subscribed. One trick I can think of is start a system that doesn't allow subscribers. Suppose you trade 20 contracts: Open a position for 20 contracts and at the exit signal close 19, and hedge the remaining one. In that way all statistics...
  12. N

    Random Trade Generator

    Monte Carlo simulations in Excel are straightforward: - use the RAND() function to generate randomness - use the Data Table functionality to generate multiple runs for any statistic that is calculated over random data
  13. N

    Technical Analysis, from a Quant's Perspective

    Andrew Lo's papers might be of your interest, e.g. http://web.mit.edu/alo/www/Papers/techanal.html
  14. N

    What happened to gold at 6:39EST (13 nov)?

    From Oanda: "XAU/USD Spike Correction The XAU/USD price movement at approximately 06:39 EST this morning on both FXTrade and FXGame was due to incorrect pricing we received: it was a spike, and we are presently correcting any transactions which were incorrectly executed. The users of all...
  15. N

    Pictures of your trading stations

    also note the DEC LA36 DECwriter II Terminal in the front. http://www.columbia.edu/acis/history/la36.html
  16. N

    Forex Free?

    Any progress on this?
  17. N

    Multiple systems at the same time

    I've been reading this thread with great interest and have a question about correlations. Are there any concerns that correlations might be sensitive to the frequency chosen (e.g. days, weeks, months etc.)? In theory it's possible to have a negative correlation using daily returns, but a...
  18. N

    Statistics Question

    I think if you're just interested if the strategy has a positive expectancy, you don't need to build a distribution of the trades. You just sample from the actual returns/trade.
  19. N

    Statistics Question

    Ok, here's an example how to do #1 in excel. Enter your returns/trade in column B. You'll find a random sample of these trades (40 in the current sheet) in column G, and its average in column I. The monte carlo runs are shown (100 in the current sheet) in column L, using Excel's Datatable...
  20. N

    Statistics Question

    #1: You can do this is easily with any programming language or statistical package. Excel could work as well. #2: I think there must be off-the-shelf backtesting software that allows this kind of monte carlo simulation. You can also re-program your strategy using a general programming language.
Back
Top