Hi all,
I suppose there's a lot of talk about indicators simply because you have to
start
somewhere - and there's more people starting than continuing. And indicators
are fun - like cloudbusting - you can imagine seeing lots of things that aren't
really there.
Agreed, RV has prepared us well, I didn't know how much you knew about
it ( which is a lot more than me! ) but I was trying to caution you. I figured
I
will be asking lots' of questions, so I should repay in kind - the sooner the
better.
Not sure why he is not popular. The information carried in his books is
awesome and offers much. I like his writing, but I don't believe it is
particularly lucid - maybe this is the reason. In Van Tharps book, he talks
about Money Management, but says little about RV apart from an appendix
mentioning "you should read him". Which is quite poor if you really believe that
MM is such a big part of it.
You are right, the opt f maths is easy, the reason the spreadsheets are a hassle
is not the maths or the concepts - it's the data entry I need for tracking my
results -
I'm an empirical sort of a guy . RV certainly does not require that much!. I
track
all my trades in a number of intraday markets, but also the trades I should
have
taken. Trading lets' me be flexible and spend a time with my family - kids and
parents who both need a bit of care - so life dictates that some trades are
missed.
So be it, but I do have to catch up on the stats. Hopefully, some work in AB
will
take care of this.
The "N=30" is from RV's first book. I think he states that it is the minumum
required
to describe a normal curve. I was curious about that since it's rather well
known now
that the normal curve ain't what traders deal with. The first book was a long
time ago...
I'll take your word on the difference in sampling error over different
timeframes. I
haven't kept the same track of "moving f" over the EOD trades - so I have to
admit guilt for making that assumption.Thank you.
I did a lot of empirical testing to see how 'f' would pan out over time though.
It's one
thing to "know" what the biggest loss will do to you, it's another to "feel" it.
I have big hopes for what AB can do for me - I was a contract 'C' programmer
for a number of years so the thought of programming doesn't faze me.
Unfortunately
I got into it through necessity rather than via a formal entry so I don't have
a lot of
the maths background that other university types do.
Anyway, soon as i finish this email I'm gonna send some off money to the
wonderful folk at AB...I'll just have another quick look at this
Mandelbrot-Hurst thingy.....
Cheers,
RZ
________________________________
From: brian_z111 <brian_z...@yahoo.com>
To: amibroker@yahoogroups.com
Sent: Tuesday, 5 May, 2009 2:44:36 PM
Subject: [amibroker] Re: testing multiple systems simultaneously
> The optimal portfolio is a fascinating topic.
I spent so much time on entries and systems etc. I am all panned out on that
topic (except for clip [arbitrage] trading which is a different game
altogether).
I see maximising the portfolio as the ultimate trading challenge ... how close
to 'fully invested' can we go without dying?
Day in and day out people in this forum talk about indicators etc and barely a
word about this magnificent challenge!
> I gave been using some of the Vince ideas for my futures trading
> and one of the biggest advantages it gives me is a way of keeping
> myself sane - knowing how to size a position when you are getting
> crushed or elated is important. It's great having a number to tell
> you how greedy or scared to be!
Hear, hear!
Vince is pretty clear about the fact that he is preparing us to handle the big
scares and giving us a chance to ease off the throttle before we hit the bend.
I am amazed that he is not more popular.
Over three years in this forum and barely a word about him.
> To give you an idea of where I find optimal f useful:
> - I stay well short of the optimal amount.
> - Where it does come in useful is at the portfolio level.
> - I trade a number of systems in futures:
> - intraday Index and Bonds - Asian session
> - intraday Bonds - European session bonds
> - EOD Curve ( 3Yr Bond / 10 year Bond )
> - EOD Bill Spread ( 90 day treasuries ) (This has been only
> recent so no stats yet - trades go for weeks )
>
> Not sure if you would consider these as uncorrelated, but there
> seems to be a measurable advantage so far in optimising at the
> portfolio level. Not sure if this is an artifact or something that I
> can use.
Probably not but I am only guessing.
I think that the reality of global financial markets hasn't truly sunk in yet,
even after last year .... it was bizarre at one stage watching all markets
around the globe moving in sync for a while.
Only a measurement will tell.
>I've been doing it in spreadsheets and it is a lot of work
> to get it right.
Implementing Vince et al, isn't the hardest maths going around ..... his early
books include Basic and C code for his early models.
His latest book has a couple of Java examples for his Leveraged Space Portfolio
Model (I am not up to speed on that one yet).
If there isn't AB implementations around already then it shouldn't be hard to
write them.
Eventually I will do it, and share it, if someone else doesn't beat me to it.
>I also suspect that the number of intraday trades >which
> are needed to be significant is way more than other time frames.
I don't think so.
Sample Error is a LAW - inviolable just like gravity.
I am not a mathematician, but I also find the subject fascinating.
To compensate for my maths naivity I spent a lot of time 'in the lab' using
Excel to test randomness and sample error ... I found sample error to be theory
that is water tight in the real world.
I have been nagging the forum about it ever since ... I found that I had to
perform repeated lab runs, and eyeball the end results, to convince teach
myself to believe in the power of variance .. the variance that a 50/50 fair
coin toss can produce is intuitively 'unbelievable' .
I have a little bit of interesting work (observations mainly) on propagation of
sample error and the flow on effect that has on our evaluation metrics ...
haven't posted it anywhere because it is very raw and posting/explaining takes
up a lot of time.
> The other thing I've noticed about optimal f ( at the itnraday >level at
> least ) is that it is moves around more than I expected it to. >Once
> again this could be due to what i think is greater noise at the intraday
> level.
> To give you an example, after 50 backtested trades I get on optimal
> f of 0.55. Vince suggests that 30 trades should be sufficient - but >he
> does acknowledge that opt f does move around. Anyhow, now I'm
> trading, and I get to
> - trade 80 ( 50 backtest, 30 actual ) opt f is 0.60
> - trade 109 ( 50 bt, 59 actual ) opt f is 0.36
> - trade 191 ( 50 bt, 141 actual ) f is 0.45 and finally seems to be
> getting a bit more stable.
I have been the lone voice in the (forum) wilderness ranting against the sample
error propaganda machine.
N == 30 came from academia ... 30 is around the number of data points where
certain (important to academia and sometimes relevant to traders) stats become
useful i.e. namely probability distributions that become normal.
Mathematicians are paid (salaries) to make their mistakes ... whereas we have
to pay for ours.
Plot sample error, as %, against N ... use a spreadsheet ... you will see that
if you want consistency then anything below N == 1000 is a joke.
Off the top of my head, your numbers are within the expected sample error for
low N datasets.
Optimal f is like the arithmetic mean ... it is asymptotic ... in laymans'
terms that means it is very stable with time.
Read RV again ... he is a brilliant educator and thinker .. his work is packed
with relevant issues ... some just get a mention ... work each point around and
you will see that all of the answers are there e.g. he does say that the higher
the N the better.
I am not one for making adjustments, based on short term analysis .. I wouldn't
change my system or my staking on 100 current samples.
> What is really dangerous ( or seems to be in my limited experience ) is
> that the largest losing trade can take you out when trading at optimal f.
> If you are trading a non-discretionary trading system at full f, then
> you MUST pull the plug if you see the biggest losing trade coming your
> way. I can't remember if Vince mentions it, but the backtesting/ calcuation
> of optimal f assumes that the biggest loss is in the past. Experience seems
> to indicate other wise!
Actually in the intro to one of his later books he gets really cranky about the
fact that this is the common view of his work.
He definitely said that the biggest loss is in the future.
He also definitely said that we can expect massive drawdown if we want to trade
with optimum growth (he is just a theoretician because he worked with Larry
Williams in those heady days when Williams traded up $2M in one year and down
1.9 the next (or something like that ... so presumably Willims went out and did
it, on Ralph's maths!)
I thought that just told us that we should put more effort into designing
systems with less variance in them ... plus the whole idea of non-correlated
portfolios is to smooth the variance.
(Your intraday example probably just has more variance in it than you are
assuming? or understanding? )
> I've put none of this into AB yet. I've still got the trial version! As soon
> as I can tear my self away from exploring it I will buy it!
That will be in a year or two :-)
I hope you are not in a relationship .. AB is a relationship buster for sure!
--- In amibro...@yahoogrou ps.com, i cs <ics4...@... > wrote:
>
> Hi guys,
>
> The optimal portfolio is a fascinating topic.
>
> I gave been using some of the Vince ideas for my futures trading
> and one of the biggest advantages it gives me is a way of keeping
> myself sane - knowing how to size a position when you are getting
> crushed or elated is important. It's great having a number to tell
> you how greedy or scared to be!
>
> To give you an idea of where I find optimal f useful:
> - I stay well short of the optimal amount.
> - Where it does come in useful is at the portfolio level.
> - I trade a number of systems in futures:
> - intraday Index and Bonds - Asian session
> - intraday Bonds - European session bonds
> - EOD Curve ( 3Yr Bond / 10 year Bond )
> - EOD Bill Spread ( 90 day treasuries ) (This has been only
> recent so no stats yet - trades go for weeks )
>
> Not sure if you would consider these as uncorrelated, but there
> seems to be a measurable advantage so far in optimising at the
> portfolio level. Not sure if this is an artifact or something that I
> can use. I've been doing it in spreadsheets and it is a lot of work
> to get it right. I also suspect that the number of intraday trades which
> are needed to be significant is way more than other time frames.
>
> The other thing I've noticed about optimal f ( at the itnraday level at
> least ) is that it is moves around more than I expected it to. Once
> again this could be due to what i think is greater noise at the intraday
> level.
> To give you an example, after 50 backtested trades I get on optimal
> f of 0.55. Vince suggests that 30 trades should be sufficient - but he
> does acknowledge that opt f does move around. Anyhow, now I'm
> trading, and I get to
> - trade 80 ( 50 backtest, 30 actual ) opt f is 0.60
> - trade 109 ( 50 bt, 59 actual ) opt f is 0.36
> - trade 191 ( 50 bt, 141 actual ) f is 0.45 and finally seems to be
> getting a bit more stable.
>
> What is really dangerous ( or seems to be in my limited experience ) is
> that the largest losing trade can take you out when trading at optimal f.
> If you are trading a non-discretionary trading system at full f, then
> you MUST pull the plug if you see the biggest losing trade coming your
> way. I can't remember if Vince mentions it, but the backtesting/ calcuation
> of optimal f assumes that the biggest loss is in the past. Experience seems
> to indicate other wise!
>
> I've put none of this into AB yet. I've still got the trial version! As soon
> as I can tear my self away from exploring it I will buy it!
>
> Z
>
>
>
>
>
>
> ____________ _________ _________ __
> From: brian_z111 <brian_z111@ ...>
> To: amibro...@yahoogrou ps.com
> Sent: Tuesday, 5 May, 2009 12:08:28 PM
> Subject: [amibroker] Re: testing multiple systems simultaneously
>
>
>
>
>
> Have you guys read Vince .... the first book, chapter 6, "The Total Portfolio
> Approach"?
>
> He gives a very good account of how to construct an optimal portfolio using
> system diversification ... it would be quite easy to implement in AB..... his
> mathematical modelling is quite straightforward. .... basically he nets the
> (special case) period returns (as growth factor) of each system .... net
> (portfolio) geometric mean == reward and variance of the net (portfolio)
> geometric mean == risk.
>
> He also discusses Markowitz and the Capital Asset Pricing model and briefly
> shows why the 'geometric mean portfolio strategy' is a valid extension of
> those models.
>
> I'm not a Vince, or a portfolio, guru and I don't have the time for an in
> depth project at this stage ..... just some quick comments ... scanning your
> posts it seems that an understanding of the theory, or some portfolio theory,
> needs to precede the implementation ... IMO:
>
> - each system should be optimized independently first
> - systems should be standardised to a common time frame e.g. daily returns
> - correlation of the period returns of the already optimised systems should
> be measured (how are you all doing that in AB?)
> - the only optimisation that is done at the portfolio level is allocation of
> capital (this is done on an iterative basis in Vinces model).
> - I find that Money Management combined with equity curve analysis (of any
> kind) is logically flawed because Money Management varies the outcomes e.g.
>
> one trade series .... +10%,-10%;
>
> start $100 -> 110,99;// a 1% loss of capital
> start $100 -> 110, add 890 capital -> 900;//approx 10% loss of capital
> - an account can never be fully invested unless we find the HolyGrail of
> systems?
>
> Optimal f was developed specifically to minimise the drag on equity recovery
> caused by asymmetrical leverage i.e. it calculates the optimum staking to
> return equity to a postive state after the max loss has been encountered.
>
> - if you want to reopt then you will need to recalc 'all of the above again'.
>
> Out of curiosity ... are you finding it easy to come up with noncorrelated
> systems?
>
> --- In amibro...@yahoogrou ps.com, "bh.hicks" <bh.hicks@ .> wrote:
> >
> > I am basically looking for a way to have AmiBroker run multiple systems
> > concurrently in order to examine how trading multiple non-correlated
> > strategies affect drawdowns. I think if there was a way to "name" an entry
> > condition so that stops and position sizing rules could be applied to a
> > particular entry criteria, it would be possible to do without too many
> > changes to AB architecture.
> >
> > I can already do this in excel using exported equity curves but it would be
> > nice to be able to do this internally so that the optimizer engine could be
> > exploited.
> >
> > Is anyone aware of a technique to do this and if not, is this something
> > others would find useful if integrated into a future version?
> >
> > As always - thank you.
> >
>
>
>
>
>
> Enjoy a safer web experience. Upgrade to the new Internet Explorer 8
> optimised for Yahoo!7. Get it now.
>
Enjoy a better web experience. Upgrade to the new Internet Explorer 8
optimised for Yahoo!7. Get it now.
