SystemTrader's Account Talk

[Spaf]

imported post

Supply and demand: too many cars, to few dinosaurs. People would be driving less if they lowered the cost of Viagra!

Hey, a reality check!

 

[SystemTrader]

imported post

Still 100% in the G Fund. Been there since 2 March. I've added a little commentary below.

The Mini-Rally (Did it end Friday? We'll Have to See...)

It would've been nice to catch a small piece of the mini rally last week. However, my system isn't designed to jump in and out of all the small rallies we have in a longer-term downturn. As I've said before, trying to catch such small moves (known as "swing trading" or "position trading") is quite difficult with a 2-day lag.

I've tested short-term trend-following and countertrend (buying on dips) systems that attempt to profit from such smaller moves. Neither has outperformed buy-and-hold for an extended period. The two-day lag is the primary killer. I'd be much more willing to position trade with something like ETFs that can be traded intraday. But I don't expect the TSP folks to add intraday-traded ETFs to the mix anytime soon.

Having said that, I know others here are using systems to play these shorter-term moves. Maybe they'll prove me wrong. I certainly haven't tested all the possible ways of short-term trading the TSP. I guess time (and FundSurfer's tallies) will tell. It should be interesting.

Indicators

My overall stock timing system employs nine timing models. One of these modelsis based on the Nasdaq's volume.You may be thinking, "great, butwe don't have a Nasdaq-based fund in the TSP." That's true, butincreases in the Nasdaq's volumeoftenlead to a rise in thegeneral stockmarket--and vice versa for volume decreases andfalling markets.I mention this because this particular model flashed a sell signal in early March, andhasan extremely weak reading right now.Even during the uptrend fromMon-Thur of this week, Nasdaq volume was weak, and is a long way from showing any real conviction.

Incidentally, if you used this volume-based model asyour only timing tool for the S&P 500 since 1987, your annualized return would be over 3% higher than a buy-and-hold approach.

Oil, Interest Rates and the U.S. Dollar

I don't think I can improve upon Dr. John Hussman'sassement on thesethree market movers & shakers. (Published tonight in his weekly column at http://www.hussmanfunds.com/wmc/wmc050411.htm).

"a number of key concerns for the market – interest rates, oil prices, the U.S. dollar – all improved, but broad market action failed to respond. Financials, bank stocks, corporate bonds and other interest-sensitives failed to improve in response to the rally in Treasuries, transportation stocks fell apart despite the pullback in oil prices, neither market breadth, leadership or other internals have exhibited strength, and so on...Given that these measures have been among the most important concerns for investors in recent months, the failure of stocks to respond to their improvement implies that there is more negative information than meets the eye. A clearer, but slightly improper way of saying this is 'if stocks can't rally on lower rates, lower oil and a stronger dollar, what happens if they deteriorate again?'.

Seasonality

A popular andeasymarket timing strategy is to buy at the beginning of November and sell at the end of April. This is also known as "sell in May and go away" (until November, that is). For thesecond half of the 20th Century, almost allthe market's gainsoccured in the Nov-Apr time frame. As Yale Hirsch of the Stock Trader's Almanacfame put it:

"A compounding investment of $10,000 in the S&P 500 Index from November through April, starting Nov. 1, 1950, and ending April 30, 2000, yielded a whopping $363,353 gain over almost 50 years. From May through October over that same period, the investment returned a "puny" $11,574 gain."

Interestingly, the folks who've really popularizedthis method (Hirsch and market timer/newsletter writerSy Harding) now use modified versions. Both require a secondary technical indicator to confirm market entry/exit and may invest within a longer/shorter time frame (e.g., from October to May if their technical indicator "signals" them to do this.) This may be because the simpler"Sell in May"system has been rather off and on since it got so much attention in 2000-2001.

I've heard one analysttheorize thatinvestors now anticipate the November buy and1 Maysell by buying early (mid-October) and selling early (mid-April). There's notenough evidenceyet to make a strong case for this yet, but we did seethe market decline rather severely at the end ofApril 2004. It will be interesting to see if this plays out again towards the end of this month.

Until next time,

John

 

[coolhand]

imported post

Thanks for the great post ST. I hear ya and I'm onboard with those thoughts. I expect to be mostly G by Thursday (or sooner). Hopefully on a positive note :?.

 

[SystemTrader]

imported post

Another quote from Dr. Hussman's article (linked in my comments from Sunday below). This quote isfrom the legendary Richard Russell of Dow Theory fame. (http://www.dowtheoryletters.com/)

"...in a bear market, whatever can go wrong will go wrong."

Very appropriate quote for the market actionthis week...especially as several stock indices fell below their yearly lows earlier today.

ST


[/b]

Still 100% in the G Fund. Been there since 2 March. I've added a little commentary below.

The Mini-Rally (Did it end Friday? We'll Have to See...)

It would've been nice to catch a small piece of the mini rally last week. However, my system isn't designed to jump in and out of all the small rallies we have in a longer-term downturn. As I've said before, trying to catch such small moves (known as "swing trading" or "position trading") is quite difficult with a 2-day lag.

I've tested short-term trend-following and countertrend (buying on dips) systems that attempt to profit from such smaller moves. Neither has outperformed buy-and-hold for an extended period. The two-day lag is the primary killer. I'd be much more willing to position trade with something like ETFs that can be traded intraday. But I don't expect the TSP folks to add intraday-traded ETFs to the mix anytime soon.

Having said that, I know others here are using systems to play these shorter-term moves. Maybe they'll prove me wrong. I certainly haven't tested all the possible ways of short-term trading the TSP. I guess time (and FundSurfer's tallies) will tell. It should be interesting.

Indicators

My overall stock timing system employs nine timing models. One of these modelsis based on the Nasdaq's volume.You may be thinking, "great, butwe don't have a Nasdaq-based fund in the TSP." That's true, butincreases in the Nasdaq's volumeoftenlead to a rise in thegeneral stockmarket--and vice versa for volume decreases andfalling markets.I mention this because this particular model flashed a sell signal in early March, andhasan extremely weak reading right now.Even during the uptrend fromMon-Thur of this week, Nasdaq volume was weak, and is a long way from showing any real conviction.

Incidentally, if you used this volume-based model asyour only timing tool for the S&P 500 since 1987, your annualized return would be over 3% higher than a buy-and-hold approach.

Oil, Interest Rates and the U.S. Dollar

I don't think I can improve upon Dr. John Hussman'sassement on thesethree market movers & shakers. (Published tonight in his weekly column at http://www.hussmanfunds.com/wmc/wmc050411.htm).

"a number of key concerns for the market – interest rates, oil prices, the U.S. dollar – all improved, but broad market action failed to respond. Financials, bank stocks, corporate bonds and other interest-sensitives failed to improve in response to the rally in Treasuries, transportation stocks fell apart despite the pullback in oil prices, neither market breadth, leadership or other internals have exhibited strength, and so on...Given that these measures have been among the most important concerns for investors in recent months, the failure of stocks to respond to their improvement implies that there is more negative information than meets the eye. A clearer, but slightly improper way of saying this is 'if stocks can't rally on lower rates, lower oil and a stronger dollar, what happens if they deteriorate again?'.

Seasonality

A popular andeasymarket timing strategy is to buy at the beginning of November and sell at the end of April. This is also known as "sell in May and go away" (until November, that is). For thesecond half of the 20th Century, almost allthe market's gainsoccured in the Nov-Apr time frame. As Yale Hirsch of the Stock Trader's Almanacfame put it:

"A compounding investment of $10,000 in the S&P 500 Index from November through April, starting Nov. 1, 1950, and ending April 30, 2000, yielded a whopping $363,353 gain over almost 50 years. From May through October over that same period, the investment returned a "puny" $11,574 gain."

Interestingly, the folks who've really popularizedthis method (Hirsch and market timer/newsletter writerSy Harding) now use modified versions. Both require a secondary technical indicator to confirm market entry/exit and may invest within a longer/shorter time frame (e.g., from October to May if their technical indicator "signals" them to do this.) This may be because the simpler"Sell in May"system has been rather off and on since it got so much attention in 2000-2001.

I've heard one analysttheorize thatinvestors now anticipate the November buy and1 Maysell by buying early (mid-October) and selling early (mid-April). There's notenough evidenceyet to make a strong case for this yet, but we did seethe market decline rather severely at the end ofApril 2004. It will be interesting to see if this plays out again towards the end of this month.

Until next time,

John

 

[SystemTrader]

imported post

(I wrote most of this earlier today, when the Dow was down about 60-100 points…but I didn't have a chance to send it. Boy was I shocked to see how much more the market dropped in the afternoon!)

I'll repeat Richard Russell's quote from my post yesterday: "...in a bear market, whatever can go wrong will go wrong." I may sound like a lonely Old Testament prophet--or maybe just a broken record--but I'll offer my $0.02 again today:

Trying to find the bottom in a market like this is very dangerous. Being highly invested in stock funds at this point is playing with fire, IMHO. Trying to guess where the bottom of a market will be is like trying to catch a falling knife.

While I've been out of stocks since early March, I'll admit we had a consolidating (though declining) period where you may have been able to buy on dips--if that's your style. We're past that now. We've taken out the yearly low, and it's much safer to let the market tell us when the bottom has set in and when the tide is turning back up.

While I follow a completely systematic, rule-based method, I know many folks use discretionary approaches (i.e., they look at indicators, charts, the news, etc., then use their judgement to make decisions.) If you do this, I'd recommend carefully noting high and low points in the stock market. These can be high/low points for the month, quarter, year, etc. Move out of stock funds--or at least reduce your allocations in them--when the market drops past a significant low point. Do the same when it can't move beyond a significant high point (e.g., in mid-March and early April). If you don't think the market will drop any lower, just remember what happened from 2000-2002.

Don't get me wrong--I'm not a perma-bear or a doom and gloomer.But I don't think I'll be in stocks any time soon--my timing system is very biased against stocks right now. I expect to beback in stocks some time during the year, but it will probably be much later. I'm sure I'll miss some mini-rallies, but Idon't think they're worth playing in the current market climate.



Regards,

ST

 

[coolhand]

imported post

John,

You've had my attention all along since you've been on this board and I've learned quite a bit from you as well as others. Your advice is quite sound. Without experiencing these last 6 months for myself however, I would have found it impossible to fully appreciate anyone's investing opinion. The experience has given me a valuable frame of reference from which I canmore accurately judge the value and soundness of any opinion offered.

Although you've only been posting a relatively short time,I understand enough now that I would have no problem blindly following your moves (or Spaf (I know that makes him nervous )). Tom is more bullish, but his reasoning is very good.

I have to wonder at this point, what the "right" move would beon Monday. If we are near the bottom and I get out now, I take a loss and perhaps miss a significant upward move. Or, as you've pointed out, we could be in a similar downward spiral like 2000-2002. But the way I see it, the market can't keep falling day after day without some movements back up. I think that is where I am at right now. Watching and waiting.

Interest rates are dropping again, which means easier access to money for Joe Sixpack. We may get another mini housing boom as well. The fed can't very well allow this economy to tank, not with the deficits we are currently running. That means the fed may back off interest rate hikes, which could inject some positive sentiment back into this market (some more liquidity wouldn't hurt either and I expect that to happen). Also, oil and commodities in general are now falling as the perceptionis that theeconomy is slowing, which translates intoa declining risk of inflation moving forward.

I don't know where all this is going, but there is a lot of market manipulation going on. I think it may take a little while for things to get back on track, but I do believe there are many reasons for the market to reverse coursewithin a couple of months.

Thanks for your .02. Please don't stop giving it.

 

[SystemTrader]

imported post

Sidenote: Thanks for the comments, Coolhand. I'll try toreply sometime this weekend.

New allocation: 100%F Fund

IFT date:18Apr 05

Moving to 100%F Fund as of COBMonday,18 Apr 05.


Since Inception Returns (starting atfunds' closing prices on 17 Dec 04):

SystemTrader: +5.11%

G Fund: +1.50%

F Fund: +0.67%

C Fund: -3.70%

S Fund: -5.53%

I Fund: +1.88%

20% in each fund: -1.04%

2005 Returns (starting atfunds' closing prices on31 Dec 04):

SystemTrader: +1.18%

G Fund: +1.31%

F Fund: +0.48%

C Fund: -5.19%

S Fund: -7.26%

I Fund: -1.94%

20% in each fund: -2.52%


Allocation History:
(Note: each date = the IFT date)

12/17/04 (start): 100% I Fund

1/18/05: 100% F Fund

1/25/05: 100% I Fund

2/02/05: 100% F Fund

2/22/05: 100% G Fund

2/24/05: 100% I Fund

3/2/2005: 100% G Fund

4/18/2005: 100% F Fund

~John

 

[SystemTrader]

imported post

coolhand wrote:

I have to wonder at this point, what the "right" move would beon Monday. If we are near the bottom and I get out now, I take a loss and perhaps miss a significant upward move. Or, as you've pointed out, we could be in a similar downward spiral like 2000-2002. But the way I see it, the market can't keep falling day after day without some movements back up. I think that is where I am at right now. Watching and waiting.

Hi Coolhand,

There will have toa rally--and itmay come soon. However, my system just isn't designed to jump in and out of upswings in very bearish conditions like we have now. When you design a market timing/asset allocation system, you have to first decide the time frame and types of moves you want to "catch": either long/intermediate term like mine, or short-term.

I think a longer term system works better considering the TSP's two-day lag. But there's another reason to be very careful if you're trying to catch a rally in a bearish market. Large, consecutive dailydrops (and gains)in the market happen much more often than they should, statistically speaking. This is known as the"fat tails" effect. Remember the bell curve from Stats 101? Well, if you plot a curve like that for a financial market, it will have an unusual number of extreme events,producing"fat tails" at each end of the curve. And these extremely bearish (or bullish) days tend to "cluster" together, meaning there are often multiple days ofvery bearish (or bullish) behavior.

Traders have known this for years. That'swhy they want to be on the right side of the trend. But this is now being recognized by the academic finance community, too.I did a quick search on Googleand found a paperthat makes this point with some terribly awkward academic language:

Therefore, the anomalously large amplitude of the drawdowns can only be explained by invoking the emergence of rare but sudden persistences of successive daily drops, with in addition correlated amplification of the drops. Why such successions of correlated daily moves occur is a very important question with consequences for portfolio management and systemic risk, to cite only two applications. (Source: http://alf.nbi.dk/~johansen/Papers/outlier5revised.pdf)

We don't know when these "daily drops" will end, or if they'll reemerge after a short "relief" rally. Therefore, Ithinkit's risky to try to catch a rally right now.

John

 

[coolhand]

imported post

Thanks for the reply ST. I'm liking your investing strategy (system) a whole lot about now...:^.

 

[SystemTrader]

imported post

We hit another yearly low today. The gains fromlast fall are all but gone for abuy-and-hold approach.

I did a little checking, and the S&P 500 was 1118.15 at the 20 April 2004 close...exactly one year ago. It's had a whopping 1.73% gain since. The real kicker is that the S&P 500 closed at 1434.54 on 20 April 2000...exactly five years ago. It will take more than a 26% gain to reach that point again, and more than 30% to hit the S&P's peak in March 2000. Ouch.

ST

 

[SystemTrader]

imported post

Still 100% in the F Fund.

 

[coolhand]

imported post

You know John, when you went 100% "F" fund on 4/18 Ihad a hard timeseeingbond yields drop by very much.I mean, how canbond prices rise again given the market data in the last month or so? Two weeks ago I was pretty sure the 10yr note was going to stay above 4.4% and slowly start to rise. Right now it sits at 4.14%. It's dropped almost a half percent inshort order. How crazy is that?

Asof today you are up a nickle in a week and a half. Considering how ugly this market is I am impressed.

 

[MR CARLO]

imported post

http://www.behaviouralfinance.net/

Hello, sys trader

Above is a Link I think you might find interesting. Also I have been looking into a system type approach to investing. A Very complicated task. Being an un-experienced investor I have much to learn. So in the near future I hope you may help me by answering some questions. Happy investing, and keep up the good work.

 

[SystemTrader]

imported post

carloc wrote: [/b]

http://www.behaviouralfinance.net/

Hello, sys trader

Above is a Link I think you might find interesting. Also I have been looking into a system type approach to investing. A Very complicated task. Being an un-experienced investor I have much to learn. So in the near future I hope you may help me by answering some questions. Happy investing, and keep up the good work.

Hi Carloc,

Interesting link: lots of information at that site. Actually, systematic trading doesn't have to be complicated. Some systems are very simple, while others are truly complex. What makes systematic trading unique is that you have precise rules for buying and selling, and you follow them 100% of the time. I wrote a good bit earlier tonight on systematic trading here (check the lastfew posts):

http://www.tsptalk.com/mb/forum11/1578.html

 

[SystemTrader]

imported post

I'm responding to Timer's question (posted in my Account folder) here. Yes, I'm still 100% in the F Fund. I know we're having something of a rally right now, but my system hasn't given a "buy stocks" signal, so I'm staying put.

Thanks,

John

 

[Timer]

imported post

Sorry about the mis-post. I knew better. Thanks John for the clarification.I know you are following a system.

 

[SystemTrader]

imported post

Still 100% in the F Fund. It was going very well until Friday...Ouch!

The bond timingsystem still has a "buy" signal, but it's weakening. Itcouldmove toa sell signal soon.

As I ran the numbers after Friday's close, I wassurprised to see that the stock timing system isclose to a buy signal. It could even happen this week...or not. The market appears to be at an important phase. It couldcontinue up and re-challenge the early April highs--and then the yearly highs after that. Or maybe this is just an unusually strong mini-rally and we'll soon sink back into the doldrums. We should find out soon.

ST

 

[SystemTrader]

imported post

Well...it happened. The stock timing system said to buy after today's close.

New allocation: 100%I Fund

IFT date:10May 05

Moving to 100%I Fund as of COBTuesday,10May 05.


Since Inception Returns (starting atfunds' closing prices on 17 Dec 04):

SystemTrader: +5.22%

G Fund: +1.78%

F Fund: +0.87%

C Fund: -0.63%

S Fund: -2.14%

I Fund: +2.62%

20% in each fund: +0.50%

2005 Returns (starting atfunds' closing prices on31 Dec 04):

SystemTrader: +1.27%

G Fund: +1.59%

F Fund: +0.67%

C Fund: -2.17%

S Fund: -3.94%

I Fund: -1.23%

20% in each fund: -1.01%


Allocation History:
(Note: each date = the IFT date)

12/17/04 (start): 100% I Fund

1/18/05: 100% F Fund

1/25/05: 100% I Fund

2/02/05: 100% F Fund

2/22/05: 100% G Fund

2/24/05: 100% I Fund

3/2/2005: 100% G Fund

4/18/2005: 100% F Fund

5/10/2005: 100% I Fund

~John

 

[pyriel]

imported post

Good timing...

 

[coolhand]

imported post

Yeah, here I am thinking he's getting ready to jump into the "C" fund and he plunges into the "I"...LOL

Makes some sense though. Dollar is hitting resistance and it seems we've found a bottom. Trade report tomorrow. Doubt it's good news, especially with the 10-Yr note back to a low level again.Just don't seethe dollarmaking much more headway. Back down it goes.