Is there a place for fundamentals in trend trading?
-
- Contributing Member
- Posts: 8
- Joined: Mon Jul 14, 2003 3:22 pm
- Location: Minneapolis
Is there a place for fundamentals in trend trading?
I am a fairly new trader, and trend trading is a new concept to me. My question is: Do fundamentals have a place in this system? Or does one simply choose a stock randomly, and apply the Turtle Rules? What I am trying to figure out, is what makes a stock making a new 55 day high(or a 20 day high) a better choice than the next? What would be something to look for that would make one stock a better selection over another?
Thank you,
Grand Cayman
Thank you,
Grand Cayman
-
- Roundtable Knight
- Posts: 101
- Joined: Mon Apr 28, 2003 11:07 am
- Location: Zimbabwe
You could construct two portfolios, run the Turtle System on each, then see whether the backtest trading results are (or are not) significantly different.
Let portfolio P be the 100 stocks of the Nasdaq 100.
Let portfolio Q be the 100 stocks whose listings in the newspaper, immediately follow the listings of the Nasdaq 100. For example, Apple Computer ("AAPL") is in the Nasdaq 100. The next stock in the newspaper is Atlantic American Corp ("AAME"). So you put AAME into portfolio Q. Continue this for the other stocks in the Nasdaq 100.
Now you've got a random portfolio of stocks ("Q") and a well known portfolio of stocks ("P"). Turtle-ize them both. Do they perform about the same? If so, what does this mean? If not, and if P outperforms Q, what does that mean?
You can start with http://screening.nasdaq.com/heatmaps/heatmap_100.asp
Let portfolio P be the 100 stocks of the Nasdaq 100.
Let portfolio Q be the 100 stocks whose listings in the newspaper, immediately follow the listings of the Nasdaq 100. For example, Apple Computer ("AAPL") is in the Nasdaq 100. The next stock in the newspaper is Atlantic American Corp ("AAME"). So you put AAME into portfolio Q. Continue this for the other stocks in the Nasdaq 100.
Now you've got a random portfolio of stocks ("Q") and a well known portfolio of stocks ("P"). Turtle-ize them both. Do they perform about the same? If so, what does this mean? If not, and if P outperforms Q, what does that mean?
You can start with http://screening.nasdaq.com/heatmaps/heatmap_100.asp
-
- Roundtable Knight
- Posts: 344
- Joined: Sun Apr 20, 2003 2:42 pm
- Location: Silicon Valley / San Jose, CA USA
- Contact:
bondtrader answered the randomization portion, so i'll answer the fundamental portion. There are two types of fundamental information. You can backtest something like:
latest quarter p/e below 20 + 55 day breakout
latest quarter price/book < 2 + 20 day breakout
It's very hard to backtest something like this:
buy FDA drug approval + 55 day breakout
analyst recommendation + 20 day breakout
If you trade with fundamental or partial fundamental information, it gets very hard to backtest data that is not easily verifiable. I didn't like the bond market for the last 6 months because interest rates were low historically. I waited until there was a breakout confirmation in July before I started shorting the long bond. That was a partial fundamental, but mostly system/mechanical trade. The fundamentals gave me a little bit more confidence in the breakout, but not much. It's just another trade.
Edward
latest quarter p/e below 20 + 55 day breakout
latest quarter price/book < 2 + 20 day breakout
It's very hard to backtest something like this:
buy FDA drug approval + 55 day breakout
analyst recommendation + 20 day breakout
If you trade with fundamental or partial fundamental information, it gets very hard to backtest data that is not easily verifiable. I didn't like the bond market for the last 6 months because interest rates were low historically. I waited until there was a breakout confirmation in July before I started shorting the long bond. That was a partial fundamental, but mostly system/mechanical trade. The fundamentals gave me a little bit more confidence in the breakout, but not much. It's just another trade.
Edward
William O'Neill - IBD
William O'Neill ("How to make Money in Stocks" and "Investors Business Daily") is probably the best known teacher of blending fundamental with technical analysis.
His CANSLIM method looks for stocks with rapidly rising earnings, high margins, high relative price appreciation and a strong industry, then jumps on the stocks when they hit a few predetermined entry points, with a 7% sell stop.
There is obviously much, much more to this, but you can find out more at http://www.investors.com
-Doug
P.S. I have no interest or affiliation with his company.
His CANSLIM method looks for stocks with rapidly rising earnings, high margins, high relative price appreciation and a strong industry, then jumps on the stocks when they hit a few predetermined entry points, with a 7% sell stop.
There is obviously much, much more to this, but you can find out more at http://www.investors.com
-Doug
P.S. I have no interest or affiliation with his company.
Re: William O'Neill - IBD
Hey there,drm7 wrote:William O'Neill ("How to make Money in Stocks" and "Investors Business Daily") is probably the best known teacher of blending fundamental with technical analysis.
His CANSLIM method looks for stocks with rapidly rising earnings, high margins, high relative price appreciation and a strong industry, then jumps on the stocks when they hit a few predetermined entry points, with a 7% sell stop.
... ...
I have a question: Is it possible to backtest the CANSLIM method? I've found that in order to backtest the method, one should find the fundmental data involved. Is there any sites that provide such type of data?
I am pretty new to mechanical systems development, so please forgive me if my question is stupid.
-
- Full Member
- Posts: 17
- Joined: Sun Aug 17, 2003 8:03 pm
Re: William O'Neill - IBD
Not a stupid question at all. Most vendors do not provide access to fundamental data through their API -- I have never understood that. For example, TradeStation has something called the Research Tool, where one can examine the fundamental data: valuation, profitability, capitalization, etc. The problem is that a programmer (through EasyLanguage) cannot access these research data to implement a strategy such as CANSLIM, or any strategy that combines fundamental and technical data. Seems like something like this would be simple to implement but inertia is high...Hey there,
I have a question: Is it possible to backtest the CANSLIM method? I've found that in order to backtest the method, one should find the fundmental data involved. Is there any sites that provide such type of data?
I am pretty new to mechanical systems development, so please forgive me if my question is stupid.
WS
Fundamentals
I personally trade currency, not stocks..but these rules should also apply
Using Fundamentals as a Secondary Indicator
1) Reaction to fundamental news
If the trend is going one way, and fundamental news comes out that seem to contradict the trend. However, the trend remains unbroken and just shakes off the bad news. In this case, I increase my position
2) When fundamental news comes out and confirms my technical indicators, I also increase my position.
3) When the trend is weakening, and fundamental news comes out contradicting my trend. I stay in my positoin, but would start to gradually scale out and take some profits.
Using Fundamentals as a Secondary Indicator
1) Reaction to fundamental news
If the trend is going one way, and fundamental news comes out that seem to contradict the trend. However, the trend remains unbroken and just shakes off the bad news. In this case, I increase my position
2) When fundamental news comes out and confirms my technical indicators, I also increase my position.
3) When the trend is weakening, and fundamental news comes out contradicting my trend. I stay in my positoin, but would start to gradually scale out and take some profits.
Hi WaveSeeker, Thanks for your reply. Seems that it's hard to find a satisfactory backtester to test fundamental and technical aspects of trading all together.
However, I wonder how, and from what way, the investors evaluate their strategies without analyzing their past performance. So, there's another questions arosed - Does it means that investors never backtest their trading rules? And moreover, for some hedge funds with "event driven" style, in which a lot of fundamental data has to be involved, how do they fine-tune their trading strategies? Do they build some proprietary statistical tools in order to do this job?
However, I wonder how, and from what way, the investors evaluate their strategies without analyzing their past performance. So, there's another questions arosed - Does it means that investors never backtest their trading rules? And moreover, for some hedge funds with "event driven" style, in which a lot of fundamental data has to be involved, how do they fine-tune their trading strategies? Do they build some proprietary statistical tools in order to do this job?
-
- Roundtable Knight
- Posts: 118
- Joined: Tue Apr 15, 2003 7:44 pm
- Location: Arizona
-
- Roundtable Knight
- Posts: 1842
- Joined: Tue Apr 15, 2003 11:02 am
- Contact:
There are plenty of quantitative fund managers that do extensive research and back-testing using fundamental data.
In fact, Max Zavanelli, besides being an International Chess Master, runs Zavanelli Portfolio Research which does quantitative analysis of fundamentals out of Lithuania. They publish research reports based on back-tested simulations of portfolio decision algorithms.
There is a real dearth of reliable fundamental data so building a reliable database is often a very labor intensive process. He found that he could get smart people in Lithuania that would help build and maintain a very solid database of various fundamental data for much lower cost than in the U.S.
He has built what is one of the most complete historical fundamental databases in the world. Unfortunately, his data, like all data of this quality is proprietary and not for sale.
As the topic's question, there are definitely ways to incorporate fundamental data into mechanical trading systems that enhance purely priced-based systems. It's not easy to do, but it does work if you look at the right data, I've seen it myself.
- Forum Mgmnt
In fact, Max Zavanelli, besides being an International Chess Master, runs Zavanelli Portfolio Research which does quantitative analysis of fundamentals out of Lithuania. They publish research reports based on back-tested simulations of portfolio decision algorithms.
There is a real dearth of reliable fundamental data so building a reliable database is often a very labor intensive process. He found that he could get smart people in Lithuania that would help build and maintain a very solid database of various fundamental data for much lower cost than in the U.S.
He has built what is one of the most complete historical fundamental databases in the world. Unfortunately, his data, like all data of this quality is proprietary and not for sale.
As the topic's question, there are definitely ways to incorporate fundamental data into mechanical trading systems that enhance purely priced-based systems. It's not easy to do, but it does work if you look at the right data, I've seen it myself.
- Forum Mgmnt
Sources like Value Line can be used to generate databases of fundamentals.
Quant recipes like magic formula investing produce outsize returns with 1-year holding periods. These are "value" approaches.
"Values" typically get that way from a falling price. The mean reversion occurs over long time frames.
By screening for value stocks, you then find a pool of stocks that, in theory, are more "primed" for an uptrend than most stocks are. You could place them on a watchlist and wait for an uptrend, then apply position sizing and trailing exits. This should outperform the value approach (which outperforms the indices).
Quant recipes like magic formula investing produce outsize returns with 1-year holding periods. These are "value" approaches.
"Values" typically get that way from a falling price. The mean reversion occurs over long time frames.
By screening for value stocks, you then find a pool of stocks that, in theory, are more "primed" for an uptrend than most stocks are. You could place them on a watchlist and wait for an uptrend, then apply position sizing and trailing exits. This should outperform the value approach (which outperforms the indices).
Here you can backtest fundamental strategies:
www.portfolio123.com
Tim, did you ever consider to enlarge the power of Trading Blox Builder so as it is possible to integrate fundamental Rules as an additional factor?
www.portfolio123.com
Tim, did you ever consider to enlarge the power of Trading Blox Builder so as it is possible to integrate fundamental Rules as an additional factor?
-
- Site Admin
- Posts: 9015
- Joined: Tue Apr 06, 2004 1:41 pm
- Location: Boston, MA
- Contact:
-
- Roundtable Knight
- Posts: 2071
- Joined: Fri Apr 25, 2003 3:33 pm
- Location: London
- Contact:
-
- Roundtable Knight
- Posts: 2071
- Joined: Fri Apr 25, 2003 3:33 pm
- Location: London
- Contact: