[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [amibroker] Re: PositionSize / Capital



PureBytes Links

Trading Reference Links

Ed and Daniel:

I tried the code below but without the Min part, using an old system of mine. However, I made the following changes:

PosQty = 20; // very large ala TJ's suggestion
SetOption("MaxOpenPositions",posqty);
rsk = -1; // 1% risk
PositionSize = Max(-20,rsk*C/StopLoss)); // where stoploss was 2*ATR(5). I was using a 1-day delay system, so I didn't need to ref the Close back a day.

Also, I made sure I set positionsize shrinking to ON and the trade size to no more than 1% of volume.

Examining the trade report. I observed I was fully invested from beginning to end, and most if not all of the time, the volatility-based positionsizing was used rather than the 5 equal equity positions. So, TJ's idea worked like a charm. It allowed me to fully invest my capital, albeit with more positions than I had comtemplated (there were times when as many as 11 positions were put on at the same time!).

Unfortunately, the overall results compared rather poorly to the equal equity model, where the code was:

PosQty = 5:
SetOption("MaxOpenPositions", posqty);
PositionSize = -100/posqty;

I do not understand why the volatility-base positionsizing compares so poorly to the equal equity positionsizing. The system tested was a very short term system, whose average trade duration was only about 2-3 days. Perhaps in such a short time period, volatility doesn't enter into the picture that much. If any of you guys have a longer system to test the above on, let me know the results you get. Perhaps that's the answer. Does volatility only play a role for longer term systems?

Al Venosa

ed nl wrote:
This way you can use a range: Maximum 20% minimum 10% of equity:
 
rsk = -2; // 2%
PositionSize = Min(-10,Max(-20,rsk * Ref(C,-1
) / stopLoss));
 
In practice it most of the time it probably either uses 10% or 20%.
 
Ed


----- Original Message -----
Sent: Sunday, December 12, 2004 3:40 PM
Subject: [amibroker] Re: PositionSize / Capital


Al & Ed,
This is exactly where I ended up yesterday (hours after my post).
When I tried it, though, I always ended up taking the 20% positions
rather than those defined by my risk. Thinking it wasn't working, I
gave up and went to bed.

But since someone else thinks this should work, obviously I need to
play with it some more.

Dan

--- In amibroker@xxxxxxxxxxxxxxx, "ed nl" <ed2000nl@xxxx> wrote:
> Al,
>
> about the part:   "Your suggestion to limit positionsize not to
exceed any more than 20% of equity may be the solution since it goes
hand in hand with the philosophy of money management. That is, do not
allow any one position to exceed, say, 10 or 15 percent of your
equity. The Turtles did that, and I think lots of traders do that,
too. So, I see nothing wrong with that. Have you coded this in AFL"
>
> I think you can solve this using:
>
> rsk = -2; // 2%
> PositionSize = Max(-20,rsk * Ref(C,-1) / stopLoss);
>
> now it will never use more than 20% of equity.
>
> About the minimum number of trades I don't know. In my system that
would be impossible because sometimes good entries just dry up and I
can't find even find 5.
>
> rgds, Ed
>
>   ----- Original Message -----
>   From: Al Venosa
>   To: amibroker@xxxxxxxxxxxxxxx
>   Sent: Sunday, December 12, 2004 3:11 PM
>   Subject: Re: [amibroker] Re: PositionSize / Capital
>
>
>   Dan:
>
>   Thanks for the ideas. You're not rambling; you're thinking, and
this discussion is healthy. Good ideas may stem from the discussion,
so by all means, keep posting.
>
>   I don't think you need a new built-in function called MinPos.
Maybe TJ came up with a solution the other day by suggesting you set
the max open positions to some large value like 10 of 15, even though
you plan to take on no more than 5 at any time. So, if you don't use
up all your equity using volatility-based positionsizing, you might
add on new positions with this approach. I haven't tested this idea
yet, but I will. The problem occurs when the opposite happens,
namely, all your equity is used up before you are able to add your
4th and 5th positions. Your suggestion to limit positionsize not to
exceed any more than 20% of equity may be the solution since it goes
hand in hand with the philosophy of money management. That is, do not
allow any one position to exceed, say, 10 or 15 percent of your
equity. The Turtles did that, and I think lots of traders do that,
too. So, I see nothing wrong with that. Have you coded this in AFL?
I'm like Yuki: good with concepts buy lousy with creative
programming.
>
>   Al Venosa
>
>   danielwardadams wrote:
>
>
>     After thinking about this some more, I think all I've described
is
>     what could be accomplished with two more built-in variables.
MinPos
>     could say you want no less than some minimum number of
positions (5
>     in my example) and MaxPositionSize could say you want to
allocate no
>     more than X% of capital to any one position (20% in my example).
>
>     Within these constraints, your actual position sizing methond
could
>     be anything you want.
>
>     I'm probably rambling .........
>
>     Dan
>
>     --- In amibroker@xxxxxxxxxxxxxxx, "danielwardadams"
>     <danielwardadams@xxxx> wrote:
>     >
>     > Al & Anthony,
>     > I've also seen the lower returns for volatility based versus
equal
>     > equity position sizing in the past and didn't know what to do
about
>     > it (assuming I wanted more positions for more
diversification).
>     >
>     > I'm not sure how one would code it in .AFL, but would the
following
>     > represent a reasonable compromise?
>     >
>     > (1) Start with an equal equity based model based on, say,  5
>     > positions (position size = -20). So each part of the pie
equals 20%
>     > of total equity.
>     > (2) Determine actual position size within each piece of the
pie
>     based
>     > on volatility based sizing. So, depending on your risk
parameter,
>     one
>     > might use only 17% of one piece of the pie, 13% of another
piece,
>     and
>     > 20%, 8%, and 11% of the other pieces.
>     > (3) Sum the used portions of the pie (in this case
17+13+20+8+11 =
>     > 69%) and see what you have left. 31% in case.
>     > (4) Allocate the remaining cash according to the equal equity
>     model.
>     > This means you get one more 20% piece of pie and only have
11% cash
>     > remaining.
>     > (5) Apply the above using your ATR based position sizing
>     recursively
>     > until your cash is minimized. So if you only are able to use
9% of
>     > the piece of pie left in (4) you take the 11% left from that
piece
>     > plus the 11% cash and you have 22% -- enough for another
position.
>     So
>     > in this case you end up with 7 positions and only 2% left in
cash.
>     > So your cash is minimized and all your positions adhere to
the ATR
>     > based position sizing.
>     >
>     > Like I say, I have no idea how to code it but intuitively it
makes
>     > sense to me.
>     >
>     > Thoughts/comments?
>     >
>     > Dan
>     >
>     > (And, yes, I'm sure I'm not the first person to think of it
so my
>     > apologies to those who have gone before).
>     >
>     > --- In amibroker@xxxxxxxxxxxxxxx, "Anthony Faragasso"
>     <ajf1111@xxxx>
>     > wrote:
>     > > Hello Al,
>     > >
>     > > You stated:
>     > >
>     > > "the lower the volatility, the lower the risk and
therefore, the
>     > smaller the positionsize for that stock. "
>     > >
>     > > Is this a correct assumption ? ...Would you want a larger
>     > positionsize on a less risk position , and a smaller position
on a
>     > more volatile one ?
>     > >
>     > > Anthony
>     > >   ----- Original Message -----
>     > >   From: Al Venosa
>     > >   To: amibroker@xxxxxxxxxxxxxxx
>     > >   Sent: Saturday, December 11, 2004 7:53 AM
>     > >   Subject: Re: [amibroker] PositionSize / Capital
>     > >
>     > >
>     > >   Ed,
>     > >
>     > >   I, too, have confirmed many times with backtesting what
you
>     > report, viz,, that positionsize = -x gives better performance
>     results
>     > than using volatility-based MM positionsizing. The non-MM
code I've
>     > used in the past is:
>     > >
>     > >   posqty = Optimize("posqty",5,2,10,1); // no. of stocks
active
>     at
>     > any given time
>     > >   PositionSize = -100/posqty; //equal equity model
>     > >
>     > >   I think I know what the problem is, but I have not as yet
>     figured
>     > out how to solve the problem with AFL. If you use the MM-
based
>     > positionsize statement as we have discussed (equal volatility
>     model),
>     > i.e., PositionSize = -1 * C/StopAmt, and examine the
tradelist, you
>     > will likely discover that, often, not all 5 stocks are active
all
>     the
>     > time. In other words, either you have idle capital earning
nothing
>     or
>     > you have fewer active stocks than you want. Why is this?
Because
>     some
>     > stocks, which might not be as volatilie as others, use up
more of
>     > your capital to initiate a position than a more volatile
stock.
>     > Consequently, your capital is used up before you have a
chance to
>     > enter into your 4th or 5th stock. Instead of having 5 open
>     positions,
>     > you might only have 3 because of this. Checking positionsize
>     > shrinking doesn't help because you'll discover you might have
tiny
>     > positions in your 5th stock. The fewer stocks you have, the
less
>     > diversified you are, and therefore the more risky your
portfolio.
>     The
>     > more risk, the higher the DDs. This problem cannot happen
with the
>     > equal equity model since all positions are equal in size, by
>     > definition.
>     > >
>     > >   One possible way around this might be to increase your
margin
>     so
>     > that equity is expanded enough to allow full funding of all
>     > positions. But, again, this also increases your risk. Another
way
>     > might be dynamically setting your risk to fit the volatility
of
>     each
>     > stock individually (the lower the volatility, the lower the
risk
>     and
>     > therefore, the smaller the positionsize for that stock).
However,
>     > this changes your model so that you no longer have equal
>     > volatility/equal risk (getting closer to the equal equity
model).
>     So,
>     > the problem remains unsolved for the moment. I have not had
time to
>     > devote to cracking this problem yet, but some day I hope to
do
>     this.
>     > If you have any ideas, I'm all ears.
>     > >
>     > >   Al Venosa
>     > >
>     > >
>     > >   ed nl wrote:
>     > >     Thanks for your effort Al. It is very clear,
>     > >
>     > >     In one of my earlier posts I posted
>     > >
>     > >     // money management block
>     > >     stopLoss = Ref(bbb*ATR(20),-1);
>     > >     // trade risk
>     > >     tr = IIf(Buy,(stopLoss / BuyPrice),stopLoss /
(ShortPrice +
>     > stopLoss));
>     > >     // renormalisation coefficient
>     > >     rc = 0.02 / tr;
>     > >     // positionsize
>     > >     PositionSize = rc * -100
>     > >
>     > >
>     > >     it actually gives the same result as your:
>     > >     PositionSize = -2.0 * IIf(Buy,BuyPrice,ShortPrice) /
stopLoss
>     > >     except for short positions. Exact the same it would be
if I
>     > use: tr = IIf(Buy,(stopLoss / BuyPrice),stopLoss /
(ShortPrice));
>     > >
>     > >     Unfortunatelly I do not get better results this way.
Using
>     just
>     > a simple PositionSize = -10 still gives somewhat better
results.
>     > >
>     > >
>     > >
>     > >     rgds, Ed
>     > >
>     > >
>     > >       ----- Original Message -----
>     > >       From: Al Venosa
>     > >       To: amibroker@xxxxxxxxxxxxxxx
>     > >       Sent: Saturday, December 11, 2004 4:19 AM
>     > >       Subject: Re: [amibroker] PositionSize / Capital
>     > >
>     > >
>     > >       ed nl wrote:
>     > >
>     > >         Al,
>     > >
>     > >         but how do you implement the risk factor now?
>     > >
>     > >         ed
>     > >       Ed:
>     > >
>     > >       Let us suppose you have established your risk as 1%
(i.e.,
>     > the maximum you are willing to lose on a trade). Let us also
>     suppose
>     > your initial equity is $100,000. So, if the stock you buy (or
>     short)
>     > goes down by the amount based on your system, you lose only
$1000,
>     > keeping you in the game. Now, let us say you defined your
>     volatillty-
>     > based stop in terms of 2*ATR(20), which you incorrectly
assigned to
>     > the variable TrailStopAmount. I say 'incorrectly' because the
>     > TrailStop in AB was designed to mimic the Chandelier exit,
which is
>     > basically a profit target type of stock (it hangs down like a
>     > chandelier from the highest high since the trade was
initiated, if
>     > long). I don't think you want the TrailStop to be your money
>     > management stop. Rather, the MM stop is the max stoploss,
defined
>     as:
>     > >
>     > >       StopAmt = 2*ATR(20);
>     > >       ApplyStop(0,2,StopAmt,1);
>     > >
>     > >       So, if your stock declines by 2*ATR(20) from your
entry,
>     you
>     > exit with a 1% loss. Let's take an example. Stock A is
selling for
>     > $40/share. It's ATR(20) is $1/shr or 2.5% of 40. Your stop
amount
>     is
>     > 2*ATR(20), which is $2/shr. How much stock do you buy? You
simply
>     > divide your risk, $1000, by 2*1, which is 500 shares. This
amounts
>     to
>     > an investment of $40/shr * 500 shrs or $20,000. All of this
can be
>     > coded in one simple line of AFL plus the 2 lines above
defining the
>     > MM stoploss:
>     > >
>     > >       PositionSize = -1 * BuyPrice/StopAmt;
>     > >
>     > >       where -1 is 1% of current equity (0.01 * 100,000 or
$1000),
>     > BuyPrice = $40/shr, and StopAmt is 2. Keep in mind that a
negative
>     > sign means 1% of CURRENT equity, which means compounded
equity, not
>     > just a constant initial equity of $100,000. If you carry
through
>     the
>     > above math with your renormalization coefficient notation,
you wind
>     > up with the exact same answer.
>     > >
>     > >       One more thing. When you place your order, assuming
you are
>     > trading with EOD data, you do not know what the buyprice is
until
>     you
>     > buy the stock, which is the next day. So, what most traders
do is
>     > base their positionsize on the closing price of the night
before
>     the
>     > entry. Therefore, to place an order in the evening to be
filled in
>     > the morning at the open, your positionsize statement would
actually
>     > be:
>     > >
>     > >       PositionSize = -1 * C/StopAmt;
>     > >
>     > >       where C is the closing price on the night before you
buy.
>     So,
>     > if you use the code SetTradeDelays(1,1,1,1), then the above
formula
>     > is OK. However, if you use SetTradeDelays(0,0,0,0), then you
have
>     to
>     > ref the C back a day.
>     > >
>     > >       This is probably more information than you were
asking
>     about,
>     > but I hope it helps.
>     > >
>     > >       Cheers,
>     > >
>     > >       Al Venosa
>     > >
>     > >
>     > >
>     > >
>     > >   Check AmiBroker web page at:
>     > >   http://www.amibroker.com/
>     > >
>     > >   Check group FAQ at:
>     > http://groups.yahoo.com/group/amibroker/files/groupfaq.html
>     > >
>     > >
>     > >         Yahoo! Groups Sponsor
>     > >               ADVERTISEMENT
>     > >             
>     > >       
>     > >       
>     > >
>     > >
>     > > ------------------------------------------------------------
------
>     --
>     > ----------
>     > >   Yahoo! Groups Links
>     > >
>     > >     a.. To visit your group on the web, go to:
>     > >     http://groups.yahoo.com/group/amibroker/
>     > >      
>     > >     b.. To unsubscribe from this group, send an email to:
>     > >     amibroker-unsubscribe@xxxxxxxxxxxxxxx
>     > >      
>     > >     c.. Your use of Yahoo! Groups is subject to the Yahoo!
Terms
>     of
>     > Service.
>
>
>
>
>
>     Check AmiBroker web page at:
>     http://www.amibroker.com/
>
>     Check group FAQ at:
http://groups.yahoo.com/group/amibroker/files/groupfaq.html
>
>
>
>
>
>   Check AmiBroker web page at:
>   http://www.amibroker.com/
>
>   Check group FAQ at:
http://groups.yahoo.com/group/amibroker/files/groupfaq.html
>
>
>         Yahoo! Groups Sponsor
>               ADVERTISEMENT
>             
>       
>       
>
>
> --------------------------------------------------------------------
----------
>   Yahoo! Groups Links
>
>     a.. To visit your group on the web, go to:
>     http://groups.yahoo.com/group/amibroker/
>      
>     b.. To unsubscribe from this group, send an email to:
>     amibroker-unsubscribe@xxxxxxxxxxxxxxx
>      
>     c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.





Check AmiBroker web page at:
http://www.amibroker.com/

Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html





Check AmiBroker web page at:
http://www.amibroker.com/

Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html





Check AmiBroker web page at:
http://www.amibroker.com/

Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html



Yahoo! Groups Sponsor
ADVERTISEMENT
click here


Yahoo! Groups Links