Page 1 of 2

How do you determine the "trend" aside of a 20-day

Posted: Wed Jun 25, 2003 11:33 am
by TradingCoach
I would like to build a stock system, where the short portfolio only goes short when the trend is down of that stock! Aside from using the much talked about BO (turtle) technique, what good measurement could one have programmatically to safely say the trend is ....(Up/Down)?
I would not be opposed to go long when there is a counter trend move to the upside...

Posted: Wed Jun 25, 2003 7:01 pm
by ForrestGump
G'day TC,

In the interests of keeping it simple, what about a simple (or exponential) MA?

Most of my trading to date is in equities and equity derivatives. I generally have a 100 day SMA up on my chart and find that the change in direction of the MA corresponds to my "eyeball" calling of "up" or "down" current trend.

It crossed my mind (but didnt stop for coffee and scones) that it may be interesting to historically examine this. The 100 day that I run is just an arbitrary number (originally chosen to ensure that I spanned at least 3 months). Other numbers on other vehicles may be more appropriate.


Hmmm...thinking a little more. Is there a relationship between VOLATILITY and the number of days that should be used for the MA to mark the trends. Another thought to place on the memory stack.

Does html work here?

<img src="http://members.optusnet.com.au/~omni1/sfe.png">

Posted: Wed Jun 25, 2003 7:05 pm
by ForrestGump
Does html work here?

<img src="http://members.optusnet.com.au/~omni1/sfe.png">
Apparently not.

How about this?

Image

Moving averages

Posted: Thu Jun 26, 2003 8:07 am
by Vince
I have heard that some people use a combination of two moving averages, such as a 20,5 ma, but I can't be sure. Maybe it's one should try and let us know the results.

Posted: Thu Jun 26, 2003 12:06 pm
by gms
ForrestGump:
That incrediblecharts chart for SFE... I was under the impression their charting software does not include US stocks. Does it?

Posted: Thu Jun 26, 2003 6:05 pm
by ForrestGump
G'day gms,
That incrediblecharts chart for SFE... I was under the impression their charting software does not include US stocks. Does it?
SFE is the ASX code for Sydney Futures Exchange (a stock I happen to be watching but that's another story). Incrediblecharts do not yet include US stocks but they now have available "500 US indices". Unfortunately that is only available in the Premium (paid for) version (quie cheap, though).

G'day Vince,
I have heard that some people use a combination of two moving averages, such as a 20,5 ma, but I can't be sure.
Moving average crossover signals are quite commonly used, I believe. I have used some very tight crossovers in the past for daytrading of warrants (who said I lack courage?). From memory, I used something like the 3 day crossing the six day. I actually achieved quite good results but it was a lot of work simply because to make a profit on it you had to watch depth for the entire day to pick intra-day entry and exit. On reflection, it was probably this that skimmed the profits rather than the supposed mechanics of the crossover signal.

I have never formally tested the use of MAs as entry/exit signals but I think that you are right that it is possibly worth testing. I always retain in my mind the oft stated maxim that there are many workable entry/exit signals but it is the money/risk management that makes it profitable or not.

I am still (slowly) writing my testing software and it will not be hard at all to include the ability to test the use of MA signals. Just don't expect anything real soon

MA' simple but effective

Posted: Fri Jun 27, 2003 1:05 am
by billpritjr
It is refreshing to see people talk about MA's.

I trade a group of 15 stocks that I have researched, going back 3 years, and found particular MA sets that work very well with them.

In almost all cases, the MA set will be 5/20, 5/30, 7/35, or 10/35. I use 10/75 for overall market condition on the S&P 500.

Basically, if the MA on stock XYZ crosses to the short, I short sell it, no thought given. This is because my back testing on stock XYZ has shown that any MA crossovers that occur are extremely reliable. So, I will take a position.

Since I am watching the same 15 stocks, which by the way are all high-volume, "big name" companies, both NASDAQ and NYSE, I am not up all night reading charts and crunching data.

BILL

Posted: Fri Jun 27, 2003 6:43 am
by Vince
Yes! You could even try using more sensitive settings and using the crosses to compound - could be interesting.

Short Portfolio

Posted: Fri Jun 27, 2003 8:15 am
by bagherra
I think it would be tough to devise a functional short portfolio based on downtrends in the stocks. Shorts of the high PE earnings momentum variety typically react VERY SUDDENLY, not gradually, to bad news. Their information flow/price reaction is different from longs.

Also, I'm doubtful that equity shorts are technically tradeable because WHY a stock goes up or down is very important on the short side. Stocks can go up for stupid reasons, driven by uninformed buying. If it goes up because of a press release that you know to be specious, then why cover? I'm not saying one should use knowledge of 'why' a stock goes up as a reason not to cover if it goes significantly against you, but on balance fundamental knowledge probably can and should affect the decision making process.

Lao Tzu said: "momentum longs, fundamental shorts".

Re: Short Portfolio

Posted: Fri Jun 27, 2003 9:56 am
by TradingCoach
I don't know...most stock shorts get into trouble are unhedged. Portfolio shorts, holding long correlated stocks (based on fundemental and trend) can profit I think. Maybe the short portfolio should have a differernt risk percentage. i.e. long risk %2 short risk 1/2% (or whatever I picked this numbers out of my hat..)
Also long stock trades should also be entered counter trend and shorts only trendwise....
bagherra wrote:I think it would be tough to devise a functional short portfolio based on downtrends in the stocks. Shorts of the high PE earnings momentum variety typically react VERY SUDDENLY, not gradually, to bad news. Their information flow/price reaction is different from longs.

Also, I'm doubtful that equity shorts are technically tradeable because WHY a stock goes up or down is very important on the short side. Stocks can go up for stupid reasons, driven by uninformed buying. If it goes up because of a press release that you know to be specious, then why cover? I'm not saying one should use knowledge of 'why' a stock goes up as a reason not to cover if it goes significantly against you, but on balance fundamental knowledge probably can and should affect the decision making process.

Lao Tzu said: "momentum longs, fundamental shorts".

SMA

Posted: Wed Jul 09, 2003 1:51 pm
by Jester
I use 3 SMA's. This helps me filter sidways markets.

Jester
:P

Posted: Sun Jul 20, 2003 3:15 pm
by ES
TRADING COACH- you want to sell and pitch do it some where else. its easy to pitch bullshit like most folks selling crap to folks that think they can get rich quick. You want to assist people then do it for free because you have agood heart and your returns will come back you you ten fold. If this is how your earn your gelt then i pity you.

Posted: Mon Jul 21, 2003 10:19 am
by Bernd
:wink:

Posted: Tue Jul 22, 2003 1:14 am
by Forum Mgmnt
babyturtle, a couple of things:

First, while I've noticed tradingcoach's propensity to market his wares elsewhere (i.e. on other forums) and especially via his email newsletter, he's been pretty respectful of our standards here on this forum.

His post in this particular thread for which you rather strongly attacked him, it seems to me, was a direct response to Hiramhon's post rather than something he initiated. TradingCoach asked a simple question and did not start this thread by attempting to "hawk his wares".

Further, while it is natural that one's opinions may have been formed based on the overall behavior of an individual in many other venues, I'd like to consider only what someone has actually done here on this forum. So in some respects, we all get amnesty for the indiscretions committed elsewhere as long as we behave here.

Second, I don't believe that tradingcoach's behavior here nor in this particular thread constitutes blatant commercialism and therefore I don't believe he deserved the level of attack you wielded.

I understand your perspective, I certainly don't want this to become a forum about vendor's selling themselves. However, at the same time, many of us are engaged in for profit enterprise and I don't want to discourage the participation of the professionals, whether they be system developers, brokers, CTAs, CPOs or hedge fund operators.

I also happen to believe that it is fair and reasonable to charge for services offered. I personally do a lot free of charge, I also charge real money for some things, and this is my choice. I don't think we should be obligated to offer our services for free. Whether a particular fee for a particular product is reasonable and represents a good value is another matter, and a valid subject for reasonable discussion here on the forum.

I am happy that you feel so strongly about the benefits of this forum that you want to keep others from ruining it through blatant commercial plugs. I share your desire to keep it free from the craziness, spamming and blatant crap (can you say "Russian Crack Group"?) that keeps popping up and polluting other places, but I ask you to please, step back, try to give others a bit more of a benefit of the doubt, and forgiveness for transgressions that might have been committed elsewhere.

In the future, if you see a post that you think crosses the line, please email or PM me about it and I'll respond as soon as I can.

Got ADX?

Posted: Wed Jul 23, 2003 1:12 am
by JimBilly
Hi, you may want to take a look at ADX.

For downtrend: DMI- > DMI+ and ADX greater than some threshold (and optionally also rising).

For uptrend: DMI+ > DMI- and ADX greater than threshold (and optionally also rising).

You can test for ADX length, ADX threshold along with, perhaps, your entry signal parameters, using ADX as described as a condition for entry.

Actual trade entry would then require the condition and your entry method itself.

JB

Posted: Fri Jan 16, 2004 9:01 am
by Sebastian
I thought this was an interesting paper, doubly so because there were a couple of ideas with practical application.

An excerpt from the post-script:


Markets make sense. Price series are not chaotic, but are carried along on currents of underlying capital flow. As we have seen, those currents may be observed through their effect on price. Moreover, a proper reading of capital flows can lead to consistent trading success.

Skeptics hold that operations based only on observed price changes cannot succeed. Markets are moved by news, they argue, and since, by definition, news cannot be predicted (or it would not be news), price movement cannot be anticipated. It is a short step to conclude that price data are not linked and that price series follow a random walk.

Skeptics fail to take into account that price activity is also news. As we have noted, traders respond to news of price change, just as they respond to other sorts of news. By their collective response traders forge causal links between past price data and current price movement. Price data are linked because traders link them...
Full paper at:

http://www.equitypm.com/TheJanusFactor.htm


Luck to all,

Sebastian

Posted: Sun Jan 18, 2004 2:06 pm
by JT
I think that Tushar Chande's Aroon indicator is an interesting method for determing the trend.

I've coded it into TR, but haven't really spent a lot of time testing it, so I can't say how it performs in actual tests.

It's one of those indicators that's on my list of things to test, (along with about a 100 other ideas) Some day, I may actual get to it. :-)

In case anyone is intersted, you can see a pretty good description of the indicator and the formula here: http://stockcharts.com/education/glossary/Aroon.html

If your looking for alternative ways to define trend, this may be a good one to look at.

JT

MA CRosovers

Posted: Sat Jan 31, 2004 8:09 pm
by ES
The MA crossovers are absolute garbage. I ;ve only lost money with that nonesense. I ve tried everythingfrom ma crossovers to simple ma's to exp ma, macd;s stochastics, i met goerge lane he's nuts! but tall of those techincial indicatros are crap. they absolutely do not teach you the more important element of risk and money mgmnt and discipline. trust me they don't work i traded from 21 till 30 using that sh_t and i was either down or flat. I even tried bollinger bands for day trading which is absolute b.s.

I decided that a) i wanted to be profitable, consistent and not chained to my terminal constantly. A master controls his future and i dont allow the market to control my life, let alone all of those nonsensical techinical indicators. If anyone can 100% accurately tell me who will win the superbowl tomorrow and by what spread I still would not believe it as its not even remotely in the probabilty books.

Posted: Sat Jan 31, 2004 9:44 pm
by verec
Babyturtle,

there may be some truth in some of the things you say, but why such a violent tone?

Don't you think it would be more proper to "agree to disagree" than to utter filler word like "absolute garbage", "he's nuts", "indicators are crap", "trust me", "that sh_t", "absolute b.s.", "nonsensical technical indicators", "its not even remotely in the probabilty books" ...

All those may be right, but you don't back-up any of those attacks with any rationale. It sounds more like the list of the personal mistakes you went through, possibly misapplying some of the things you cite, rather than a reasoned, documented factual analysis of why indicator so and so when tested on market X, Y & Z over a period of T years gave more losing trades than some other technique ...

I could construe this post (and others you made on the TT situation) as: ''Don't ever use a hammer! It's an horrible tool! I just hurt two of my fingers with one of those stupid things the other day! Hammers are crap! Don't ever think of using a hammer! For any reason!"

That would sound preposterous, don't you think? :roll:

BTW: the break-out rule of 55 or 20 days in the Turtle System is an indicator.

Baby turtle

Posted: Sat Jan 31, 2004 11:09 pm
by billpritjr
MA crossovers do indeed work, it just depends on the type of security in question and the overall market trend. The MA crossover method in and by itself is not the Holy Grail, for stocks you need to look at 1) Overall Market trend 2) Sector Trend 3) Specific stock action itself

One reliable method is a 50-day EMA, a 1-day EMA, which is the same as price itself, if it penetrates upward, go long, downward (especially on 2+ times ave. volume, and overall market is down), go short.

Hold off on the stoch, MACD, other stuff. Slow down and return to basics. Richard Donchian, whose original "N-day" breakout was a blueprint for the turtle method, also "pioneered" MA crossovers.

There are other things such as risk management, market selection, volume action, but sometimes simple is better in the markets.

Good luck