fabijo's account talk

[fabijo]

Looks like I'll be hanging out in the G fund for at least another day. I just updated today's prices and it's still saying that G is the place to be.

 

[Birchtree]

You're looking at weird wally wednesday and the hammer may strike the S fund and the I fund without mercy.

 

[350zCommTech]

Looks like I'll be hanging out in the G fund for at least another day. I just updated today's prices and it's still saying that G is the place to be.

I've been watching you moves and I know it's too early to judge your spread sheet, but it seems too mechanical to me. What do you think about it so far. I'm working on one myself but it's very simple and involves some market fundamentals.

 

[fabijo]

I've been watching you moves and I know it's too early to judge your spread sheet, but it seems too mechanical to me. What do you think about it so far. I'm working on one myself but it's very simple and involves some market fundamentals.

You're right, it is a little too mechanical. I would like it to sit still for more than two days at a time. I know that anyone sticking to a good plan could have made better returns by only making a few trades during a year. What I would like to do for the spreadsheet is have it look long-term and short-term so that it can judge if we are getting close to tops or bottoms of trading channels. Since I already have the formulas in there to project days in advance, I can use those formulas backwards to decide if today's price is a top, a bottom, a high, a low, or just something in between.

I had some time today, but I haven't used that time for the spreadsheet. I'm actually researching different software I can include on my website that will allow people to easily share files like these spreadsheets. It would be alot easier for us to keep track of changes to these sheets and it would give people a chance to upload files that are much larger than the limits we have on TSPTalk. I don't know if anybody has any experience or knows which might be better, but I'm trying to decide between these three programs:

KnowledgeTree
MyDMS
PolDoc Document Management System

 

[fabijo]

Again, I hate doing this, but I'm going 100% I Fund. That's what the sheet says to do, but I can easily see why. It's the only fund that had a huge jump in price, which made the MACD give a positive slope. Two days ago the spreadsheet projected the I Fund to be within the following ranges:

Lowest: $21.80 (4 standard deviations)
Low: $22.00 (1 standard deviation)
Ideal: $22.07
High: $22.14 (1 standard deviation)
Highest: $22.28 (3 standard deviations)

The fund is pushing the limits of the High end, but all year it has been doing that. The worrysome part is when it pushes above the 3 standard deviations. If it does that, then in two days it will be going past $22.37.

 

[fabijo]

Here we go again with another IFT. This time it's to the F fund. Where's that Santa rally??

 

[fabijo]

Well, for anyone who likes to see the numbers by the year, I just totalled the spreadsheet returns by year. For 2003, it only has from June 2. Here are the results:

Year____G______F_______C_______S_______I____Monkey
2003____2.4____.2____16.15__24.18__26.84___30.13
2004____4.34__.83____11.20__17.84__19.17__28.17
2005____4.4___2.4____5.78___12.21__14.15__16.96
YTD_____4.84__4.77___13.87__13.20__21.63__31.19

 

[Pilgrim]

Well, for anyone who likes to see the numbers by the year, I just totalled the spreadsheet returns by year. For 2003, it only has from June 2. Here are the results:

Year____G______F_______C_______S_______I____Monkey
2003____2.4____.2____16.15__24.18__26.84___30.13
2004____4.34__.83____11.20__17.84__19.17__28.17
2005____4.4___2.4____5.78___12.21__14.15__16.96
YTD_____4.84__4.77___13.87__13.20__21.63__31.19

Do you like that monkey enough yet to put yourself on autopilot and let the monkey make the calls - for real?

 

[fabijo]

Do you like that monkey enough yet to put yourself on autopilot and let the monkey make the calls - for real?

I've been following the Monkey's every call since the beginning of this month. I really can't argue with his/her returns, even though he/she makes way too many moves.

When you say autopilot, are you referring to me making it automatically log into my TSP account and make the transfers? I'll definitely do that one day, but I would rather it do that from a web server running some PHP/MySQL. I'll be learning how to accomplish that during the upcoming year. Plus, I've got some probabilities and statistics books (with a touch of fractal theory) on my Christmas wishlist that I hope will help me fine tune some issues in the coming year.

 

[fabijo]

Made an IFT to 100% S fund today. I'm still being a guinea pig and following the Monkey's calls.

I just started reading a book from Benoit Mandelbrot called The (Mis)Behavior of Markets. So far it is pretty good. Let's see what I can learn from his fractal geometry theories.

 

[fabijo]

I just went 100% I, effective cob tomorrow.

I've been spending my spare minutes trying to use old market data in my spreadsheet. I'm going from January 1987 and used the S&P 500 and the Wilshire 4500 data from then till now. For the G and F funds, I kept them flat at the same price until June 2, 2003 where I used their actual prices. For the I fund, I am using the MSCI data from May 2001 till now.

The strange thing is that I've gotten completely different results for the past few years when I use the index data instead of the fund prices. I'm trying to figure out why. I didn't think it mattered what the prices were, because the sheet is meant to follow price movement, no matter how high or low the prices are. One way I've gotten it to show similar results is to calculate the MACD of the log of the index prices instead of the MACD of the index itself. Very strange.

I also noticed that my little probabilities inclusion does better during some years if I adjust the high and low boundaries. I might be able to fix that by projecting longer into the future than just two days. Another thought would be to have today's price checked against long term tops and bottoms. Using projected growth from those points, the Monkey could better judge if the market is approaching a top or a bottom.

I could completely revamp my methods after I am finished reading Mandelbrot's book. The chapters I have read are mostly giving an overview of modern economic theory and how it got there. He is then going to show the flaws in modern economics to introduce how a fractal approach better explains markets than the current bell curve approach. He is convincing me so far. My approach to price projections is using the methods that he says is flawed. It works for most situations, but can kill you during those times where the model is broken. Until I learn more, I'll keep tweaking what I have.

 

[airlift]

Fabijo,
For your sake, and I don't mean to be offensive, don't invest your nest egg in your tracker until you have improved its performance. Good Luck!

 

[ebbnflow]

Very sound advice, sponsor. Trading on paper has its merits, and I believe that fabijo is younger than most people here and is not averse to risk. Be that as it may, I would still like to see how his dart-throwing monkeys perform for the new year. Good luck for the new year, fabijo!

 

[fabijo]

Sponsor,
I am definitely not offended at your advice. Like ebb said, I'm young so I'm willing to take a risk. If I ever decide to stray away from the monkey, I'll try to keep posting the monkey's moves so those interested can see how he/she performs.

I've been able to tweak the monkey to beat the Wilshire 4500 till now. Some years it did worse and some years it did better. Here are the results:

From Jan 30, 1987 to December 26, 2006

S&P 500: 416.97%
Wilshire 4500: 456.11%
Monkey: 504.63%

I think that if I play around with the Monkey's rules, it'll do better. Some rules I am thinking of have to do with what type of market it is in or how close to a long term top or bottom the current price is approaching.

Ebb, thanks for the wishes.. same to you!

 

[airlift]

Only trying to be of help!

Very sound advice, sponsor. Trading on paper has its merits, and I believe that fabijo is younger than most people here and is not averse to risk. Be that as it may, I would still like to see how his dart-throwing monkeys perform for the new year. Good luck for the new year, fabijo!

 

[airlift]

Only trying to help! Happy New Year!

Very sound advice, sponsor. Trading on paper has its merits, and I believe that fabijo is younger than most people here and is not averse to risk. Be that as it may, I would still like to see how his dart-throwing monkeys perform for the new year. Good luck for the new year, fabijo!

 

[fabijo]

Whew. It's been too long since my last post here. Happy New Year, all!

I've been using historical data with the S&P 500, the Wilshire 4500 and MSCI EAFE. After seeing how this spreadsheet would've worked through all those years, with its variety of markets, I practically revamped the way I was going through the logic. It was too quick to leave the market.

Since this is a long-term account, I made sure that I would stay in the market more often. It seems it is more risky to get out of the market than to just stay in.

For the backtesting, I have S&P data starting in January 1950, Wilshire data starting January 1987, MSCI EAFE data starting May 2001, G fund and F fund data starting in June 2003.

From January 1950 to December 28, 2006:

S&P 500 : 8451.8%
Monkey: 54202.74%

The monkey for more recent years:

2006: 41.32%
2005: 18.81%
2004: 28.23%
2003: 33.89%
2002: 3.967%
2001: 12.57%
2000: -13.31%

Any time frame up until June 2003, every time the Monkey got out of the market, it got 0% gain because I did not have G fund and F fund data from before then. So it is possible that the Monkey would've done better in some down years if the F fund was available as an option.

Right now, the spreadsheet is HUGE (71.8MB). I'm currently in a Visual Basic class with college. Maybe by the end of the class, I can turn this into a standalone program instead of wasting hard drive space with a spreadsheet that repeats the same IF, THEN, ELSE statements thousands upon thousands of times.

 

[Show-me]

Fabijo,

Whew! Great work and thanks for sharing. How do ya do it with everything on you plate? Don't burn yourself out brother.

 

[ebbnflow]

Fabijo, I just saw the new tally for the new year. Your monkeys are leading the pack! Congrats.

 

[fabijo]

Fabijo,

Whew! Great work and thanks for sharing. How do ya do it with everything on you plate? Don't burn yourself out brother.

This kind of stuff just occupies my mind during those crazy moments when I begin to get stressed out about family matters. It's difficult to watch a family member suffer from mental problems, but sees nothing wrong with their irrational outlook.