Predicting randomness

[nononsense]

Stop babbling nonsense.
If you want to know about prediction in the mathematical sense, start with studying Wiener's "Optimum Linear Predictor".

After you UNDESTOOD this, come back and talk about prediction.

 

[cnms2]

Thanks.

Quote from dan05:

Hi CNMS,

I have not read all posts yet, but may be this link can help you.

http://www.predictionlabs.com

I'm testing their system and it seems to be predicted the market direction, at least Intraday.

They use System Dynamics, and Chaos Theory. Their results look promising.

Dan

 

[dan05]

Hi.

I'm not sure about "Optimum Linear Predictor", but as far as I've analyzed this company I was able to see incredible market predictions and trades.

Attached you'll be able to see some of them.

The black line is the prediction, while the red line is what finaly happened. The point where the black line separates from the red line is the prediction time.

These guys seem to be serious enough to justify spending some time reading their web site.

They have published some statistical information that seems to me that they demonstrate their system have a edge.

http://www.predictionlabs.com/papers/WPV1.1.pdf


Does anyone have the knowledge to read this paper and give an opinion? How do you analyze a trading system and predictor model ?

Thank you,

Quote from nononsense:

Stop babbling nonsense.
If you want to know about prediction in the mathematical sense, start with studying Wiener's "Optimum Linear Predictor".

After you UNDESTOOD this, come back and talk about prediction.

 

[nononsense]

Quote from dan05:

Hi.

I'm not sure about "Optimum Linear Predictor", but as far as I've analyzed this company I was able to see incredible market predictions and trades.

Attached you'll be able to see some of them.

The black line is the prediction, while the red line is what finaly happened. The point where the black line separates from the red line is the prediction time.

These guys seem to be serious enough to justify spending some time reading their web site.

They have published some statistical information that seems to me that they demonstrate their system have a edge.

http://www.predictionlabs.com/papers/WPV1.1.pdf


Does anyone have the knowledge to read this paper and give an opinion? How do you analyze a trading system and predictor model ?

Thank you,

Interesting but what does this mean if you pick out ONLY 2 DAYS for the confirming analysis?

Imagine at ET a guy validating a trading method by picking out 2 days out of a period of 2 or 3 years. That's equally convincing like Jack's!

 

[dan05]

Hi Nononsense.

I thought the same thing when browsing their past predictions.

http://www.predictionlabs.com/pastperformance.html


But, their Root Mean Square Error analysis, based on 3 years of data vrs random market predictions make me think on average they have an edge of 27% above the chance.

With respect to their trading system results, they show good numbers. I've been following their intraday trades,
at
http://www.elitetrader.com/vb/showthread.php?threadid=70418

and during the worst market days, they made a killing.
Like 500 DJIA Pts during the last 30 days.

I'm still doing research on them, but they might be on something here.

Quote from nononsense:

Interesting but what does this mean if you pick out ONLY 2 DAYS for the confirming analysis?

Imagine at ET a guy validating a trading method by picking out 2 days out of a period of 2 or 3 years. That's equally convincing like Jack's!

 

[Jesse333]

coming from a private pilot- amateur meteorologist..

Choas theory:

The only reason we can't predict the weather is we don't have all the data.

We understand all the physics of the atmosphere but we just can't accurately enough enter data to predict weather in the LONG TERM effectively.

ie.The flapping of a single butterfly's wing today produces a tiny change in the state of the atmosphere. Over a period of time, what the atmosphere actually does diverges from what it would have done. So, in a month's time, a tornado that would have devastated the Indonesian coast doesn't happen. Or maybe one that wasn't going to happen, does. (Ian Stewart, Does God Play Dice? The Mathematics of Chaos, pg. 141)
http://www.imho.com/grae/chaos/chaos.html

I'd tend to think that the market is even more unpredictable since only god knows what each of us are thinking at a certain time.
Hence there is no predictability of the market by entering it into an equation; or better yet we just don't have all the "data". We don't have it by a long shot.

 

[Cheese]

Quote from Jesse333:
coming from a private pilot- amateur meteorologist..

Choas theory:

The only reason we can't predict the weather is we don't have all the data.

We understand all the physics of the atmosphere but we just can't accurately enough enter data to predict weather in the LONG TERM effectively.

ie.The flapping of a single butterfly's wing today produces a tiny change in the state of the atmosphere. Over a period of time, what the atmosphere actually does diverges from what it would have done. So, in a month's time, a tornado that would have devastated the Indonesian coast doesn't happen. Or maybe one that wasn't going to happen, does. (Ian Stewart, Does God Play Dice? The Mathematics of Chaos, pg. 141)
http://www.imho.com/grae/chaos/chaos.html

I'd tend to think that the market is even more unpredictable since only god knows what each of us are thinking at a certain time.
Hence there is no predictability of the market by entering it into an equation; or better yet we just don't have all the "data". We don't have it by a long shot.

It ought to be self evident how far you are wrong.

Think 'boundaries'. The Dow in its upcoming Tuesday session (currently the future) is highly unlikely to finish 200 points above or below the Open. Why? Because it has done so, so rarely. Now adding more parameters how much more predictable do you think the Dow would become?

When you say 'only God knows' you are unintentionally right because He is a human invention or, if not, collectively all humans together are the Godhead incarnate. But put that aside.

What you're actually claiming, correctly, is that ahead of time you and others cannot make head nor tail of the markets and you therefore can predict the panic or confusion of yourself and others with certainty. See, prediction, again, works well.

 

[man]

Quote from libertad:

Every price movement has a reason...and a place to go..

And is definitely not random...To suggest that price movement has no reason...can only come from a fool...

i am not sure if "not random" suggests "no reason" ...

 

[trader28]

God can more accurately be called existence, all that is, pure living mind in essence, which makes humans a thought inside that mind

 

[steve46]

Quote from oddiduro:

If you flip a coin and get three heads on Mon, Tuesday, and Wednesday at noon precisely, does that mean it will happen on Thursday and Friday.

You sir, may have been fooled by randomness

Well this is really getting old..First a little slap on the wrist..

You know you could just buy a stat book and figure this out on your own..

If YOU flip a coin, one that you took from your pocket, lets say it is a nickel, and you let me inspect it, and I see that it is NOT a new coin, and you flip three heads, then I am going to bet that you flip heads again for the next couple of flips...The reason is that I KNOW that the heads side is slightly heavier than the tails..Its not much but it is enough to give me a slight edge..I will take that edge every time...

If on the other hand we have what is statisticians call a "fair coin" equally weighted on both sides. Then I am still going to bet on heads. Although the flips are independent, there is no reason to rule out a streak of heads, and in practice (I have already done this) streaks of 5 to 10 in a row are not unusual..Depending on the odds offered me, I will take that bet IF the risk to reward ratio is attractive. In other words it is all about evaluating the risk to reward vs the odds that a run of heads will continue...

Now Oddi, ordinarily you exhibit good common sense, so I have to suspect that recently you have had some medical procedure that requires you to receive pain meds. Otherwise you would have thought this through yourself...

Finally let me offer an observation...I would much rather bet on the continuation of a trend, then on the origination of one..For it is a much more attractive to bet that Thursday's and Friday's coin flip will be heads, than to bet that all flips will be heads starting on Monday.....I will leave you to figure that one out yourself...