Also a common cause I found is the difference of the malloc().. On some systems overrunning an allocated chunk just works, on other crashes with rather cryptic messages, and in some very different part of the code.. Using a good debugging version of malloc helps. I soleved all (most) of my trouble switching to C++ STL for development.. Will not get back to "C"...
You miss, sorry. This is only true for a finite variance variables. Observables in nature are not required to have finite variance - there are plenty of cases when they do not.
You position is typical for those who just took some statistic classes, but never bothered to check the fine print and understand what it means (no insult, please) But be careful when you make strong statements in public. They sound funny.
Check some references on "stable distributions" in statistics. For physical examples do a search on "Levy flights"
Also Gaussian distr. is not "normal" in a sense of its frequent occurence. I would claim that scaling, or "power law" as physicists refer to it, distribution is much more common in natural phenomena.
Do not suffer too much. That article is garbage. Read some books on related subjects instead. As an example Mandelbrot is pretty funny and many parts of his writing are accesible for a person without strong mathematics/statistics background..
The flight patterns of the wandering albatross appear to fit a special type of random motion called a Levy flight, which may also prove useful for describing aspects of heartbeat rhythms and other biological phenomena.
References:
Klafter, J., M.F. Shlesinger, and G. Zumhofen. 1996. Beyond Brownian motion. Physics Today (February):33.
Viswanathan, G.M., et al. 1996. Levy flight search patterns of wandering albatrosses. Nature 381(May 30):413.
Further Readings:
Gillespie, D.T. 1996. The mathematics of Brownian motion and Johnson noise. American Journal of Physics 64(March):225.
Hayes, B. 1994. Nature's algorithms. American Scientist 82(May/June):206.
Lavenda, B.H. 1985. Brownian motion. Scientific American (February):70.
Mandebrot, B.B. 1983. The Fractal Geometry of Nature. New York: W.H. Freeman.
Peng, C.-K., et al. 1993. Long-range anticorrelations and non-Gaussian behavior of the heartbeat. Physical Review Letters 70(Mar. 1): 1343.
Shlesinger, M.F. 1989. Levy flights: variations on a theme. Physica D 38:304.
Slade, G. 1996. Random walks. American Scientist 84(March/April):146.
Worsley, K.J. 1996. The geometry of random images. Chance 9(No. 1):27.
ASAIK - insurance companies do emply some pretty smart guys with a good background in statistics, physics and math. I know some people working in this area. They DO NOT jsut use a Gaussian distribution. This article is junk.. In particular - scaling distribution of the earthquakes and other natural phenomena is well accepted and broadly used - I do not understand what is the target audience for this article. But this is not "news for geeks"..
It is actually i big deal to try to persuade MBA types to understand a thing about stable distributions. Their first reaction "but it got infinite variance! " and every paper they have seen refer to variance of the process. They can not understand that when you use some parameter to describe a process - this parameters are not an intrinsic characteristic of the process - it is rather the charateristic of your model. So when you calculate a variance to describe a stock price trajectory - it does not mean that every model of this process has to have a finite variance. You just made a fit based on a finite observation. Gaussian fit this processes very badly indeed and a lot of "fixes" are developed to account for that. But people are reluctant to change their paradigm altogether - they did not studied that in that single business statistics class school they took. Gaussian theory is 100 years old, fractals, self-affine processes, Levy flights etc. are all too recent and not so well understood. It will take time.
This paper is still a scam anyway... Better model will not make you a better trader - there are too many other variables affecting security prices - psychology of the buyers is one big unknown. With people trading over the net so fast, effects of news of their selling behaviour is too hard to guess... Just my 0.02c...
I need to read before I submit.. Try again: but recent developments in the study of the self-affine (fractal) processes made other stable distributions used more often.. Still bad, but screw my English...
Arrgh... you missed you statistics class. T-distibution is NOT a BETTER FIT. It is a distribution of a sample mean of a Gaussian variable, when you use a sample variance instead of a true variance (when it is unkhown) When your sample grows bigger - the estimator of the variance becaomes more accurate and your t-distribution approaches Gaussian. There is nothing painful about t to use once you got a clue.
As for the Gaussian, as I mentioned above - it is a particular case of a stable distribution : with a finite variance. This is hardly news, but some recent developments in the self-affine processes made it more other stable distributions more widely known..
Yes, Gaussian is one case of the family of so called "stable" distribution - the only one with a finite variance. Stable distribution is such that if you add random variables with this distribution you will get a variable with the same distribution. - As in law of the large numbers, but without finite variance. Infinite variance means the tails of the distribution fall off slowly - it is more likely to get an event further from the mean value.
So fucking what? Big news? Hardly.
Stable distributions have a lot of applications in many areas of physics and finance. Do a literature search on "Levy flights" for examples. There was a good article on Levy flights in one recent "Nature" (IIRC) For some financial applications - check out very easily written (but for a specialist kinda useless - IMHO) Mandelbrot's "Fractals and Scaling in Finance". It has some good discussion on the subject.
Guys, you look like fools, making news out of a rather well known field. And discussing it rather childishly...
But as for SMP support, my original comment was that with PC architecture we have it HERE and NOW. It there is also some pretty good competition in this area. So Apple's bragging about "supercomputer on a slice of silicon" is what it is - irresponsible bragging.
It does not. Libertarian != Stalinist You blatantly confuse terms.
Personally, I do not like both Libertarians and Stalinists and all other obsessed people. They scare me. But, please, do not lie. Use accepted definitions from political science (I nknow Pol.Sci. is an oxymoron, but, hell, that the best we got for agreeing on terminology...)
Huh? Did not know that... Who make this boards? Why Apple do no sell them?
Anyway.. why does not Apple compare there new G4 to a 800Mhz Kriotech Athlon (which cost $2200, can use much better video cards and other hardware)
Sure G4 is a fine processor. But Apple's marketing drivel is annoying. It is so clearly exploits people without a slightest clue about computer usage, its sickening.
For some reason Xconfigurator does not offer 1600x1200. Going into XF86Config and adding it manually works just fine. xf86config selects the 1600x1200 mode, but screws up font server and wheel mouse configuration.
For some reason, on my TNT Vanta card, cursor --not the hardware one: I already used Option "sw_cursor" -- but the regular rectangular cursor in Emacs, does not show up over the plain (black) text. I have identical configuration with a machine using Matrox card, and that box shows cursor just fine. Running emacs with -cr Red option, chaneges the color of the selected letter, but does not show cursor over white space. I tried all emacs configurations from working setups to no success. I am puzzled and annoyed.
You are dumb, aren't you? Sure I have seen dual PIII boards BEFORE PIII shipped. It uses the same MB that PII. Any ETA for dual G4 from Apple, wise ass? And what about OS support for it? It will take a while. PC will have i64 and SMP 1Ghz Athlons long before that.
Slashdot Mirror
Also a common cause I found is the difference of
the malloc().. On some systems overrunning an
allocated chunk just works, on other crashes with
rather cryptic messages, and in some very
different part of the code.. Using a good debugging version of malloc helps.
I soleved all (most) of my trouble switching to C++ STL for development.. Will not get back to "C"...
Maybe you forgot about 64 bit? Sounds like it...
That's not an observation or a fact.
You miss, sorry. This is only true for a finite variance variables. Observables in nature are not required to have finite variance - there are plenty of cases when they do not.
You position is typical for those who just took some statistic classes, but never bothered to check the fine print and understand what it means (no insult, please) But be careful when you make strong statements in public. They sound funny.
Check some references on "stable distributions" in statistics. For physical examples do a search on "Levy flights"
Also Gaussian distr. is not "normal" in a sense of its frequent occurence. I would claim that scaling, or "power law" as physicists refer to it, distribution is much more common in natural phenomena.
Do not suffer too much. That article is garbage. Read some books on related subjects instead. As an example Mandelbrot is pretty funny and many parts of his writing are accesible for a person without strong mathematics/statistics background..
Ooups. If you was sarcastic, I missed it...
My apology..
Trails of the Wandering Albatross
Applying the mathematics of haphazard motion
The flight patterns of the wandering albatross appear to fit a special type of random motion called a Levy flight, which may also prove useful for describing aspects of heartbeat rhythms and other biological phenomena.
References:
Klafter, J., M.F. Shlesinger, and G. Zumhofen. 1996. Beyond Brownian motion. Physics Today (February):33.
Viswanathan, G.M., et al. 1996. Levy flight search patterns of wandering albatrosses. Nature 381(May 30):413.
Further Readings:
Gillespie, D.T. 1996. The mathematics of Brownian motion and Johnson noise. American Journal of Physics 64(March):225.
Hayes, B. 1994. Nature's algorithms. American Scientist 82(May/June):206.
Lavenda, B.H. 1985. Brownian motion. Scientific American (February):70.
Mandebrot, B.B. 1983. The Fractal Geometry of Nature. New York: W.H. Freeman.
Peng, C.-K., et al. 1993. Long-range anticorrelations and non-Gaussian behavior of the heartbeat. Physical Review Letters 70(Mar. 1): 1343.
Shlesinger, M.F. 1989. Levy flights: variations on a theme. Physica D 38:304.
Slade, G. 1996. Random walks. American Scientist 84(March/April):146.
Worsley, K.J. 1996. The geometry of random images. Chance 9(No. 1):27.
ASAIK - insurance companies do emply some pretty smart guys with a good background in statistics, physics and math. I know some people working in this area. They DO NOT jsut use a Gaussian distribution. This article is junk..
In particular - scaling distribution of the earthquakes and other natural phenomena is well accepted and broadly used - I do not understand what is the target audience for this article. But this is not "news for geeks"..
It is actually i big deal to try to persuade MBA types to understand a thing about stable distributions. Their first reaction "but it got infinite variance! " and every paper they have seen refer to variance of the process. They can not understand that when you use some parameter to describe a process - this parameters are not an intrinsic characteristic of the process - it is rather the charateristic of your model. So when you calculate a variance to describe a stock price trajectory - it does not mean that every model of this process has to have a finite variance. You just made a fit based on a finite observation.
Gaussian fit this processes very badly indeed and a lot of "fixes" are developed to account for that. But people are reluctant to change their paradigm altogether - they did not studied that in that single business statistics class school they took. Gaussian theory is 100 years old, fractals, self-affine processes, Levy flights etc. are all too recent and not so well understood. It will take time.
This paper is still a scam anyway... Better model will not make you a better trader - there are too many other variables affecting security prices - psychology of the buyers is one big unknown. With people trading over the net so fast, effects of news of their selling behaviour is too hard to guess...
Just my 0.02c...
I need to read before I submit.. Try again:
but recent developments in the study of the self-affine (fractal) processes made other stable distributions used more often..
Still bad, but screw my English...
Arrgh... you missed you statistics class.
T-distibution is NOT a BETTER FIT. It is a distribution of a sample mean of a Gaussian variable, when you use a sample variance instead of a true variance (when it is unkhown) When your sample grows bigger - the estimator of the variance becaomes more accurate and your t-distribution approaches Gaussian.
There is nothing painful about t to use once you got a clue.
As for the Gaussian, as I mentioned above - it is a particular case of a stable distribution : with a finite variance. This is hardly news, but some recent developments in the self-affine processes made it more other stable distributions more widely known..
Yes, Gaussian is one case of the family of so called "stable" distribution - the only one with a finite variance. Stable distribution is such that if you add random variables with this distribution you will get a variable with the same distribution. - As in law of the large numbers, but without finite variance.
Infinite variance means the tails of the distribution fall off slowly - it is more likely to get an event further from the mean value.
So fucking what? Big news? Hardly.
Stable distributions have a lot of applications in many areas of physics and finance. Do a literature search on "Levy flights" for examples. There was a good article on Levy flights in one recent "Nature" (IIRC) For some financial applications - check out very easily written (but for a specialist kinda useless - IMHO) Mandelbrot's "Fractals and Scaling in Finance". It has some good discussion on the subject.
Guys, you look like fools, making news out of a rather well known field. And discussing it rather childishly...
Agree. Intel has even worse drivel..
But as for SMP support, my original comment was that with PC architecture we have it HERE and NOW. It there is also some pretty good competition in this area. So Apple's bragging about "supercomputer on a slice of silicon" is what it is - irresponsible bragging.
I hope this clarifies things.
It does not. Libertarian != Stalinist
You blatantly confuse terms.
Personally, I do not like both Libertarians
and Stalinists and all other obsessed people.
They scare me. But, please, do not lie.
Use accepted definitions from political science (I nknow Pol.Sci. is an oxymoron, but, hell, that the best we got for agreeing on terminology...)
Huh? Did not know that... Who make this boards? Why Apple do no sell them?
Anyway.. why does not Apple compare there new G4 to a 800Mhz Kriotech Athlon (which cost $2200, can use much better video cards and other hardware)
Sure G4 is a fine processor. But Apple's marketing drivel is annoying. It is so clearly exploits people without a slightest clue about computer usage, its sickening.
: cat - > foo.c // code here ^D :-)
;)
Still better than vi
.. it breaks in all different modes, when it uses mode-default faces.
If 3.3.5 does not fix it, I am trading in my TNT...
..still do not show over white space.. ;(
Colored as well. That plenty sucks. Is it a TNT server problem? Bad bug: should be fixed..
... they can always clone a copy - remember Alien Resurrection? Ripley was dead for 200 years..
Edit XF86Config manually...
For some reason Xconfigurator does not offer 1600x1200. Going into XF86Config and adding it
manually works just fine.
xf86config selects the 1600x1200 mode, but screws up font server and wheel mouse configuration.
For some reason, on my TNT Vanta card, cursor --not the hardware one: I already used Option "sw_cursor" -- but the regular rectangular cursor in Emacs, does not show up over the plain (black) text. I have identical configuration with a machine using Matrox card, and that box shows cursor just fine.
Running emacs with -cr Red option, chaneges the color of the selected letter, but does not show cursor over white space. I tried all emacs configurations from working setups to no success.
I am puzzled and annoyed.
You are dumb, aren't you? Sure I have seen
dual PIII boards BEFORE PIII shipped. It uses the same MB that PII.
Any ETA for dual G4 from Apple, wise ass? And what about OS support for it? It will take a while. PC will have i64 and SMP 1Ghz Athlons long before that.
..you correct me, even if PPC can support SMP theoretically - there are no dual/quad G4 boxes offered..
..but does it support SMP?
Somehow I do not believe it will actually outrun my dual PIII 500 Dell in my simulation code.
Say nothnig about our 16 head Suns...
And on my home box PIII450+TNT2 will kick G4
butt in any game you find (if it supported on Mac,
which is a rarity).
Apple marketing drivel make me sick. "Think different", my ass. More honest would be "Do not think, just believe all the buzzwords..."