Predicting is Unavoidable

makosgu · Mar 15, 2007

Quote from Trader666:

You've exposed yourself as an ignorant poser with your own words. And please, speak for yourself. Try to get your facts straight for once. Don't say what I see or think as you've learned to do from your master, Jack, because my posts have obviously gone over your head and you haven't a clue what I'm talking about.

The interesting thing about this charade is that you are the one dismissing a paper that some of us have been able to incorporate and take the bank. Why is it that some traders can take it to the bank while some cannot? Is it fair to state that an individual may have a clue about said document given that they are taking it's information to the bank? Still ironically, how is it that when two people use the same code based on said document and wind up with opposite results? How would anyone troubleshoot the latter? By checking what was done perhaps??? And why would anyone refuse to have something that is busted fixed especially since it would allow them to take the then fixed tool to the bank!

Bottom line is that in trading, it's all about what you can take to the bank. If you can take randomness and prediction to the bank, by all means take it as far as you can. Acrary certainly did. Someone see randomness, someone else sees order in the randomness. The computer you are typing on is full of atoms that are built from subparticles that are doing all sorts of random things. How is it that a bunch of radom subparticles can be combined to do something orderly, like bond with another atom and create polymers and surfaces and solid structures that do non random things. Is it possible to see the markets this way???

Quote from Trader666:

I understand how the order execution process works and you seem to have some of the basics down. You've said that trades arrive randomly and on that I almost agree because it's usually the case but not always. But it's clearly lost on you that when it is random, the bid and ask FIFO queues are a cumulative consequence of the random arrival of trades, and that the extent to which the random arrival of further trades interacts with the queue and eventually moves price is also random (most of the time). You could easily verify this with a stochastic simulation if you knew how to run one which I doubt.

Hooray, we can agree on something. So why is it that this bit of knowledge is being taken to the bank by some and not others? I threw up one of my cheat sheets to show two things that connect two types of dots and how to relate what you see as random to a second thing that does not act randomly for me. So if you pulled down some data and ran an autocorrelation between BID/ASK pair change and those other values that were below the BID/ASK pair, you would get a value that was not "horizontally stable". Since you say that you are good with time series analysis, you shouldn't have any trouble calculating the value and seeing that it is NOT horizontally stable. If you do the same thing for the arrival time, you will find that time arrival is horizontally stable. You can go on and on on the time series analysis as you say you have done...

You'll still need to take it to the bank tho...

Trader666 · Mar 15, 2007

Of course you take that garbage (the P,V Boolean relation) to the bank.

I believe you... NOT!!!!!

You've got a lot to learn about time series analysis.

Good luck in candyland.

Quote from makosgu:

The interesting thing about this charade is that you are the one dismissing a paper that some of us have been able to incorporate and take the bank. Why is it that some traders can take it to the bank while some cannot? Is it fair to state that an individual may have a clue about said document given that they are taking it's information to the bank? Still ironically, how is it that when two people use the same code based on said document and wind up with opposite results? How would anyone troubleshoot the latter? By checking what was done perhaps??? And why would anyone refuse to have something that is busted fixed especially since it would allow them to take the then fixed tool to the bank!

Bottom line is that in trading, it's all about what you can take to the bank. If you can take randomness and prediction to the bank, by all means take it as far as you can. Acrary certainly did. Someone see randomness, someone else sees order in the randomness. The computer you are typing on is full of atoms that are built from subparticles that are doing all sorts of random things. How is it that a bunch of radom subparticles can be combined to do something orderly, like bond with another atom and create polymers and surfaces and solid structures that do non random things. Is it possible to see the markets this way???

Hooray, we can agree on something. So why is it that this bit of knowledge is being taken to the bank by some and not others? I threw up one of my cheat sheets to show two things that connect two types of dots and how to relate what you see as random to a second thing that does not act randomly for me. So if you pulled down some data and ran an autocorrelation between BID/ASK pair change and those other values that were below the BID/ASK pair, you would get a value that was not "horizontally stable". Since you say that you are good with time series analysis, you shouldn't have any trouble calculating the value and seeing that it is NOT horizontally stable. If you do the same thing for the arrival time, you will find that time arrival is horizontally stable. You can go on and on on the time series analysis as you say you have done...

You'll still need to take it to the bank tho...

makosgu · Mar 16, 2007

Quote from Trader666:

Of course you take that garbage (the P,V Boolean relation) to the bank. I believe you... NOT!!!!!

You've got a lot to learn about time series analysis.

Good luck in candyland.

According to the polls, for every 6 that aren't able to take it to the bank, 4 are taking it to the bank. Be 7, the facts remain it doesn't matter what you are doing. However, the facts are the facts. Two backtests, same original source code, same underlying documents, two opposite outcomes. How come??? How much sense does it make to refute have the busted outcome fixed to be a positive outcome that said person can take all the way to the bank??? Things that make some people go hmmm???

As for time series analysis, you won't believe this, but I use it every day (M-F) for risk analysis of an entirely different market arena (credit). Maybe you have matlab readily available. When I regularly run my stochastic models for Interest Rate risk (vasicek), Bond Recovery Rates, and Credit Spread Premiums (GARCH(1,1)). There all based on historical time series analysis. As you can imagine, such knowledge is very valuable to some people. But you wouldn't believe the price tag figures of that. Prediction is ****avoidable***. Just look at the map so that you know what the turns in the road look like.

As you can imagine, life is bitter sweet in candyland. Please keep up the great work and keep us posted on your backtesting progress!

nitro · Mar 16, 2007

Quote from Pet Shop Boys:
Do it (do it do it do it do it do it...)
Do it do it do it do it do it do it do it do it...
Do it now do it now do it now do it now do it now do it now do it now
Do it (do it do it do it do it...)

Late on Tuesday evening, such a commotion in the street
Someone broke a window, and someone's head got beat
A wave of breaking bottles, crashed across the city street
And someone got arrested, for the breach of the peace

If you wanna walk (walk walk walk walk)
Don't talk (talk talk talk talk)
Do it! (do it do it do it do it do it do it do it)

If you wanna earn (earn earn earn earn)
Learn (learn learn learn)
how to do it! (do it do it do it do it do it do it do it)

And if you wanna ride (ride ride ride ride)
don't hide (don't hide)
Do it! (do it do it do it do it do it do it do it do it)

If you wanna stay (stay stay stay)
Don't say (say say say)
Prove it! (prove it prove it prove it prove it prove it prove it)

How much longer are you gonna sit and talk to me?
You got so many problems and a split personality
You want to see a doctor before our love is tested
How much longer a man could get arrested?

If you wanna walk (walk walk walk walk)
Don't talk (talk talk talk talk)
Do it! (do it do it do it do it do it do it do it)

If you wanna earn (earn earn earn earn)
Learn (learn learn learn learn)
how to do it! (do it do it do it do it do it)

And if you wanna ride (ride ride ride ride)
don't hide (hide hide hide hide hide)
Do it! (do it do it do it do it do it do it)

And if you wanna stay (stay stay stay)
Don't say (say say say)
Prove it! (prove it prove it prove it prove it prove it prove it)

You've got your health, you've got everything, that's what my doctor said
You may not have much cash, but you've got a roof over your head
Of course I said I loved you, not just 'cause you insisted
How much longer a man could get arrested?

If you wanna walk (walk walk walk walk)
Don't talk (talk talk talk talk)
Do it! (do it do it do it do it do it do it do it)

If you wanna earn (earn earn earn earn)
Learn (learn learn learn learn)
how to do it! (do it do it do it do it do it do it)

If you wanna ride
don't hide
Do it! (do it do it do it do it do it do it)

If you wanna stay (stay stay stay stay stay)
Don't say (say say say say say say)
Do it! (do it do it)
Do it! (do it do it do it do it do it do it)

Why don't, why don't we do it now?
Do it now?
Why don't, why don't we do it now?
Why don't we do it now?

Why don't, why don't we do it now?
Why don't we do it now?
Why don't we do it now?

nitro

Gaucho MM1 · Mar 16, 2007

Quote from aeliodon:

Its funny Jack you're on record as saying: "I use 17 *leading* indicators to let me know whats coming up *next*." And then you say prediction has nothing to do with anything you do.

aeliodon yeah, that's pretty much right

Dictionary.com Unabridged (v 1.1) - Cite This Source
anÂ·ticÂ·iÂ·paÂ·tion /Â¿nˌtɪsəˈpeɪʃən/ Pronunciation Key - Show Spelled Pronunciation[an-tis-uh-pey-shuhn] Pronunciation Key - Show IPA Pronunciation
ânoun 1. the act of anticipating or the state of being anticipated.
2. realization in advance; foretaste.
3. expectation or hope.
4. previous notion; slight previous impression.
5. intuition, foreknowledge, or prescience.
6. Law. a premature withdrawal or assignment of money from a trust estate.
7. Music. a tone introduced in advance of its harmony so that it sounds against the preceding chord.

--------------------------------------------------------------------------------

[Origin: 1540â50; (< MF) < L anticipÃ¢tiÃ´n- (s. of anticipÃ¢tiÃ´), equiv. to anticipÃ¢t(us) (ptp.; see anticipate) + -iÃ´n- -ion]
Dictionary.com Unabridged (v 1.1) - Cite This Source

preÂ·dicÂ·tion /prɪˈdɪkʃən/ Pronunciation Key - Show Spelled Pronunciation[pri-dik-shuhn] Pronunciation Key - Show IPA Pronunciation
ânoun 1. an act of predicting.
2. an instance of this; prophecy.

--------------------------------------------------------------------------------

[Origin: 1555â65; < L praedictiÃ´n- (s. of praedictiÃ´) a foretelling. See predict, -ion]

âSynonyms 2. forecast, augury, prognostication, divination, projection.
Dictionary.com Unabridged (v 1.1)
Based on the Random House Unabridged Dictionary, Â© Random House, Inc. 2006.

=========================================

........and when I first read the title of this thread, I thought, this'll be a bun-fight bout nothing.......and it came true.......

anticipating and predicting are part of the same make-up for this auction......the same activity that is misdiagnosed by the herd and correctly applied by the smart money........one goes in-hand with the other in the realms of making money otherwise; without prediction of income the auction would not exist.......stop over complicating and side-tracking........all that posturing........what for? that's the real question........

Trader666 · Mar 16, 2007

How exactly are you avoiding prediction with those models? Or isn't forecasting prediction? LOL!!! And why are you using historical data, instead of only "NOW?" Maybe you should introduce your boss to the Hershey method!

The reason your backtests show different results might be that they're in error, or maybe you've added enough conditions to what I was testing to over-fit to the data, or maybe they're fiction like 3X daily range. Who knows! But I do know that I know how to backtest, and what I did was correct.

I usually don't discuss specific trades but one of my most recent is germane to this discussion, so... I trade mechanical, totally quantifiable, short term patterns that have proven themselves in all market conditions. Not sure if I mentioned this before but the reason I've looked into Jack's stuff on and off over the years is the possibility of finding another pattern or filter to add to my trading... I don't care where or from whom an idea comes, as long as it tests out. Anyway, I got a buy signal and bought LEND on Wednesday. Yesterday it was pulling back from its high so I sold it for a 5 point profit (70% of the 2 day range) even though my research indicated I should hold it overnight... big mistake that cost me about 3 more points. Which is why I don't trade seat of my pants and which is why backtesting is so important to me. Of course, I'm sure you all got 3X the range on LEND, right?

Quote from makosgu:

According to the polls, for every 6 that aren't able to take it to the bank, 4 are taking it to the bank. Be 7, the facts remain it doesn't matter what you are doing. However, the facts are the facts. Two backtests, same original source code, same underlying documents, two opposite outcomes. How come??? How much sense does it make to refute have the busted outcome fixed to be a positive outcome that said person can take all the way to the bank??? Things that make some people go hmmm???

As for time series analysis, you won't believe this, but I use it every day (M-F) for risk analysis of an entirely different market arena (credit). Maybe you have matlab readily available. When I regularly run my stochastic models for Interest Rate risk (vasicek), Bond Recovery Rates, and Credit Spread Premiums (GARCH(1,1)). There all based on historical time series analysis. As you can imagine, such knowledge is very valuable to some people. But you wouldn't believe the price tag figures of that. Prediction is ****avoidable***. Just look at the map so that you know what the turns in the road look like.

As you can imagine, life is bitter sweet in candyland. Please keep up the great work and keep us posted on your backtesting progress!

traderdragon2 · Mar 16, 2007

"Prediction is ****avoidable***. Just look at the map so that you know what the turns in the road look like."

oxymoron

Trader666 · Mar 16, 2007

You know, I'm so used to reading their contradictions that I missed that!

Quote from traderdragon2 quoting makosgu of candyland:

"Prediction is ****avoidable***. Just look at the map so that you know what the turns in the road look like."

oxymoron

makosgu · Mar 16, 2007

Quote from traderdragon2:

"Prediction is ****avoidable***. Just look at the map so that you know what the turns in the road look like."

oxymoron

Are maps predictions or are they based on paths that are already there??? EEK!

traderdragon2 · Mar 16, 2007

The problem here is that the map itself is a prediction in this case.

A map implies you can see the road ahead.

Since you cant do that in trading (you cant see into the future), your map in this case is a prediction, a best guess.

Quote from makosgu:

Are maps predictions or are they based on paths that are already there??? EEK!

Predicting is ***Unavoidable***

Predicting is Unavoidable