Techniques for Managing a Random Entry

Discussion in 'Strategy Building' started by inandlong, Nov 13, 2002.

  1. TSaimoto

    TSaimoto Guest

    Aphie started it and Gekko rode along...
     
    #41     Nov 13, 2002
  2. If you are doing a so called random entry, then why fool with TA? Why not just define the amount of exposure you wish to take as time? If you are short term oriented, maybe your desired exposure time is 30 minutes. If you are a bit longer term but still exit at end of day, maybe it is exit at end of day. If your desired exposure is 3 days, then exit after 3 days. Then you will have accurately captured the moves during that period of time (either as a loss or profit).
     
    #42     Nov 13, 2002
  3. TSaimoto,

    couple things...my example was only to illustrate a point. by no means am i using it as a real trading example. the $2 target and stop were pulled out of thin air. as for "Lose, double... lose, double... lose, double... " when did i say that? the concept i was discussing is about scaling out of losers, not doubling down on them. anyway, i hope i don't sound bitter here...i'm not. i just want to be clear.
     
    #43     Nov 13, 2002
  4. TSaimoto

    TSaimoto Guest

    Let's assume that the 2 point stop-loss is already set. So in another words, if you are wrong, you'll lose 2 points automatically.

    Now, I'm assuming that the entry was random(Something like when I woke up I suddenly had a position and for some reason I can't complain) and the market is going my way at the particular point.

    Then I'm going to play it safe by 1. Just getting out. Still, inandlong's point is more on how would you maximize the profits in a situation.

    So I'll keep myself safe and exit part of it so that I can reduce the 2 point loss to somewhere better. If it's above 2 move that the moment, even if I get hit with the stop-loss, I can break even. I'll let it ride after that and if I feel the wave has hit the top, I'll take some out also.

    Given the situation, that's the most I can do to play it safe and also try to profit out of it.

    Still, along the positive open position, if I get an opposite signal, I'll get out, if I get a positive signal, I'll pyramid the remainder of the position I would have traded without the random one.

    Let's say we want the best way to trade in a series of random trade. That's a different story.
     
    #44     Nov 14, 2002
  5. If your trading partner was a monkey, would you rather have the monkey responsible for the entry or exit?

    So when the fast moving average crosses above the slow moving average, you put a banana right next to the buy button.

    If you are using random entry in a strong uptrend, only take the buy signals. (You laugh, but it is no joke...think about it.) The entry is random, but you are trading with the trend, and that my friend is a moneymaker.
     
    #45     Nov 14, 2002
  6. Exacto profitseer, exacto!

    I was hoping you would be drawn in to participate.

    I'm reading a lot about trailing stops and randomness and the odds of this and that.

    Although I cannot argue against such a measure, ie., enter - set trailing stop and stop loss - wait for result, I would be surprised to find many people doing just that. Yet, the more I think about it, the more I like it. Talk about simple. At least you have a chance for a big move.

    The timed-exit is a little harder for me to buy in to. If one is given a position worth $50k to manage, and you get to keep the profits... setting the alarm clock for 10 minutes and then que sera sera seems like a wasted opportunity.

    Just kiddin' ya a little TriPack - but you could add that one to the Gems thread - "TriPack has posted a simple method that requires very little knowledge outside of setting a timer. He calls it the Buy Stock - Set Clock - Let it Rock Method."

    :)
     
    #46     Nov 14, 2002
  7. TSaimoto

    TSaimoto Guest

    I might be the only one doing this...

    Well, I'm just going to base my trade based on what I see on the chart only using general charts and Moving average.

    Here's a few comments:

    First, MA is the primary signal and it's a trend following indicator, in general, so that's what I'll be making. I need back testing to do this but I'll be relying on what I see on the chart(real life I wouldn't trust my visual ideas too much with writing systems)...

    2 entries... Red crossing price and Red crossing Cyan Line.

    Green Lines will be lines for exits I guess.

    So something like:

    If Red > Blue and Close > Green then Buy at Market.
    If Red < Blue and Close < Green then Sell at Market.

    ExitLong at Green
    ExitShort at Green


    I don't have the exact figures so I can't say much but I might put in a few filters in to take out trades during low volume/consolidating market condition. So:

    If Average(78, Volume) > Average(78, Volume)[1 day ago] then EntryOK = True.

    I guess I end with the entry there. Now position sizing. Depends but I'll just keep it 1% per trade. So:

    PositionSize = Risk Amount / (Previous Close - Green) * PointValue

    I'll also use Risk of Ruin based on drawdown another words lower risk after each loss. That should help with the series of losses while consolidating markets.

    I'll have the above as the basic system. Simple, don't have history testing so can't say it works but I'll go something like above.

    I'm sure I can still add something like:

    If Red > Blue and Red of Data2 > Blue of Data2 then Buy

    Well, let's say I'm long... Well, I would have exited out. But for some reason if I re-enter, I'll exit when they recross back down.

    *PS Green line is hard to see...
     
    #47     Nov 14, 2002
  8. Random in the colloquial sense and random in the statistcal sense, while similar, are markedly different. I should not have used the word random here because it elicits statistical discussion. This is not what I want.

    Lets forget the word random.

    Here is the scenario! You have been hired to manage a fund of one stock. The previous fund manager was fired. Your fund is long this one stock. The previous manager left no notes.

    Your job is to manage this fund. There is nothing about randomness that applies here. Manage the fund using TA and Risk Management techniques.

    How will you go about your job?
     
    #48     Nov 14, 2002
  9. MUChris

    MUChris

    Why don't some of you system programmers out there write a system with random entry and exit in TS or WL or something and post the results besides spitting out statistics theory that if anything, we all know does not apply to real life situaions.

    MUChris
     
    #49     Nov 14, 2002
  10. MUChris

    MUChris

    Sorry, realized my post read random exit too. I meant random entry and systematic exit.
     
    #50     Nov 14, 2002