Now we've got some big numbers to throw around, too

i-e88a953e59c2ce6c5e2ac4568c7f0c36-rb.png

Only ours are methodologically valid. It's a common creationist tactic to fling around big numbers to 'disprove' evolution: for instance, I've had this mysterious Borel's Law (that anything with odds worse than 1 in 1050 can never happen) thrown in my face many times, followed by the declaration that the odds of the simplest organism forming by chance are 1 in 10340,000,000. It's complete nonsense, of course — their calculations all ignore the reality of the actual events, assuming that everything must form spontaneously and all at once, which is exactly the opposite of how probability plays a role in evolution. It's annoying and inane, and the creationists never seem to learn…perhaps because the rubes they pander to are easily dazzled by even bogus mathematics, so they keep doing it.

We're going to have to start firing back. Doug Theobald, a long-time contributor to Talk.Origins and the Panda's Thumb, has written a very nice paper testing the likelihood that all life on earth is not related by common descent, and he comes up with some numbers of many digits to support evolutionary theory. Nick Matzke has a summary, and the story has been written up for National Geographic.

Basically, the idea is this: take a small set of known, conserved proteins that are shared in all organisms, not restricting ourselves to one kingdom or one phylum, but grabbing them all. In this paper, that data set consists of 23 proteins from 12 taxa in the Big Three domains: Bacteria, Archaea, and Eukarya. Then set up many different models to explain the relationships of these species. For instance, you could organize them into the classic single tree, where all are related, or you could model them as three independent origins, for each of Bacteria, Archaea, or Eukarya, or you could postulate other combinations, such as that Bacteria arose independently of Archaea and Eukarya, which share a common ancestor. Finally, you tell your computer to do a lot of statistics on the models, asking how likely it is that two independent groups would each arrive at similar sequences, rating each of the models for parsimony and accuracy against the evidence.

And the winner is…common ancestry, with one branching tree! This is what we expected, of course, and what Theobald has done is to test our assumptions, always a good thing to do.

More complicated permutations of these models were also tried. What if there were a significant amount of horizontal gene transfer? Would that make multiple origins of modern life more likely? He was testing models like the ones below, where the dotted lines represent genes that leap across taxa to confuse the issue.

i-d060a3501eaef6f6686404d4ac24ba00-origin_models.jpeg

The answer here is that they don't. These models can also be evaluated by statistical methods, and the best fit is again the one on the right, with a single ancestral root. People might recall the infamous "Darwin was wrong" cover from New Scientist—well, these results say that New Scientist was wrong, the existence of extensive horizontal gene transfer does not negate the fact of common descent.

So what's the big number? There are lots of them in the paper, since it evolves many comparisons, but Theobald distills it down to just the odds that bacteria have an independent origin from Archaea and eukaryotes:

But, based on the new analysis, the odds of that are "just astronomically enormous," he said. "The number's so big, it's kind of silly to say it"--1 in 10 to the 2,680th power, or 10 followed by 2,680 zeros.

One in 102680? Hey, aren't those odds a little worse than Borel's criterion of one in 1050?

Stay tuned to the Panda's Thumb. Apparently, once he finishes up the trifling business of wrapping up a semester's teaching, Theobald will be putting up a synopsis of his own and answering questions online.


Theobald D (2010) A formal test of the theory of universal common ancestry. Nature 465(13):219-222.

More like this

It's nice when the numbers used to support a point are a result of something other than pullitoutofyourassism.

By Rev. BigDumbChimp (not verified) on 14 May 2010 #permalink

I've been reading up on the history of mathematics and the dude who first really systematised probability calculations used 0.001 as his cut-off. Anything happening with 0.999 likelihood was in his words "morally certain". I rather like that expression.

By Sili, The Unkn… (not verified) on 14 May 2010 #permalink

I've long said that something like that should be done. Actually, I wrote a simple program for calculating just a few of the many needed antecedents of various organisms, which quickly ran into high probabilities that evolution must have occurred. But it was far from being comprehensive (I'm not much of a programmer).

So it's good, but the "answer" will always be that magic "could do" all of that, at the whim of the magician/god. Which would be true of any of nature's regularities, of course, an extremely obvious point that magic believers often willfully, always stupidly, ignore.

Glen D
http://tinyurl.com/mxaa3p

By Glen Davidson (not verified) on 14 May 2010 #permalink

But, based on the new analysis, the odds of that are "just astronomically enormous," he said. "The number's so big, it's kind of silly to say it"--1 in 10 to the 2,680th power, or 10 followed by 2,680 zeros.

Or about the odds that a creationist will be swayed by these calculations.

I like it. But it seems to basically come down to parsimony. As we know, Occam's Razor is a dull instrument. I think to make inroads, the story will have to be spiced up with virgins and aliens.

"Simples of organisms forming by chance." Huh. I like that.

What's the chance of water forming by chance, in the presence of oxygen and hydrogen? Or iron oxide, in the presence of iron and water? That must be, like, a brazillion to one!

Chemical reactions follow certain rules. In specific situations, those reactions are inevitable. What's so hard to grasp about that?

By nigelTheBold (not verified) on 14 May 2010 #permalink

What if there were a significant amount of horizontal gene transfer? Would that make multiple origins of modern life more likely?

This is a trick question because: horizontal gene transfer only makes sense in the context of a common genetic code, and a common genetic code is itself evidence of a single origin.

By Reginald Selkirk (not verified) on 14 May 2010 #permalink

That must be, like, a brazillion to one!

I am so stealing that. :-)

I think to make inroads, the story will have to be spiced up with virgins and aliens.

Ah! Like the Evony ads. Good idea.

By nigelTheBold (not verified) on 14 May 2010 #permalink

For comparison it's estimated that there are about 1080 particles in the universe. Take that number and put it to the power of 33.5 and you get 102680.

By Feynmaniac, Ch… (not verified) on 14 May 2010 #permalink

I understand why the tree on the right (the correct tree) is show the way that it is, but does anyone have good links to papers that show this correct tree with branches that did not follow on to survive? That is one that shows how the branching we know of today is not the only branching, just what survived over time? I think that more aptly illustrates what evolutionary processes actually did. That is to say that there were many branches that simply died or failed to survive long enough to get to modern day or even to a time where they could leave fossil evidence. Branches which we are more likely to find as successful on other planets or space bodies.

One problem with this approach is that Theobald ignores the mechanistic 'model' that creationists use. They don't think it all happened by chance. They think these sequences were smithed by Adam's hand, guided by his dad, or something. (Hey, were Adam and Jesus brothers? Or clones? ... Meh, why start asking questions now -- I'll never get it.)

Anyway, this is somewhat similar to their calculations of chance for evolution -- they don't allow for selection guiding the process. Oh well, serves them right. Maybe the ones who figure this out can be convinced that the creationist calculations are at least as invalid for testing evolution.

This model doesn't account for the simplest explanation of shared proteins: goddidit, using a set of uniform building blocks, because that's what He Chose To Do (after all, it was a pretty busy first week, so it only makes sense that, like GM's less intelligent designs of the 70's through the 90's, a lot of components would be shared...)

By Doktor Zoom (not verified) on 14 May 2010 #permalink

Creationists aren't going to be swayed by this, they'll just say that God created every organism similarly.

“But,” the creatidiots will insist, “my God is a small god!”

Dr. Sagan, brilliant mind that he was, had them pegged to a fare-thee-well.

Cheers,

b&

--
EAC Memographer
BAAWA Knight of Blasphemy
``All but God can prove this sentence true.''

By Ben Goren (not verified) on 14 May 2010 #permalink

Creationists aren't going to be swayed by this, they'll just say that God created every organism similarly.

It's not a matter of mere similarity. It's a matter of creating organisms with every appearance of common descent, to the degree of certainty demonstrated by these statistics. Creationists can take that line, and I suppose they will, but it amounts to making their god a cosmic trickster.

Maybe before he started planting fossils everywhere, Satan dabbled in bioinformatics.

They already say the universe was made just to look old when it's really young

Creationist headline:

Atheist Scientists Admit Independent Origin a Possibility

By Feynmaniac, Ch… (not verified) on 14 May 2010 #permalink

Sorry to repost this comment, but it's worth a look for the laughs.

http://www.theintercessors.com/index.php

~~You can enter your "prayer" and then see the other ones that have been left.
Oh and don't attempt to use proper punctuation,for some reason it won't take it. I'm not kidding.

Seems there are a few people that post the same prayers almost everyday. One girl seems to be getting it on with "David" on the weekend and wants our prayers that it will keep on going, lol.~~

@ #1

I rather like "pullitoutofyourassism". So much easier than thinking, learning, testing, and questioning. Evidence! If only it could fit in your ass.

Actually, what I really like is the "ism" attached to "pullitoutofyourass". Hahaha. :)

Ry I started to laugh when reading that site, but it's really fucking sad.

I mean the prayers themselves are sad, but the fact people put this on a website thinking that some supernatural being is going to actually help them out is even sadder.

By Rev. BigDumbChimp (not verified) on 14 May 2010 #permalink

1 in 10 to the 2,680th power, or 10 followed by 2,680 zeros.

Hmm, am I misunderstanding something, or shouldn't that be 1 followed by 2680 zeros?

By matthew.james.neil (not verified) on 14 May 2010 #permalink

Only ours are methodologically valid. It's a common creationist tactic to fling around big numbers to 'disprove' evolution: for instance, I've had this mysterious Borel's Law (that anything with odds worse than 1 in 1050 can never happen) thrown in my face many times, followed by the declaration that the odds of the simplest organism forming by chance are 1 in 10340,000,000. It's complete nonsense, of course — their calculations all ignore the reality of the actual events, assuming that everything must form spontaneously and all at once, which is exactly the opposite of how probability plays a role in evolution. It's annoying and inane, and the creationists never seem to learn…perhaps because the rubes they pander to are easily dazzled by even bogus mathematics, so they keep doing it.

Creationists keep making the mistake becasuse they were badly designed.

By jcmartz.myopenid.com (not verified) on 14 May 2010 #permalink

Fascinating stuff, and a very elegant approach, to try and calculate how likely it is that there wasn't common ancestry !
Do we know what proteins these 23 were, and what they do ?

By Rorschach (not verified) on 14 May 2010 #permalink

@matthew.james.neil

You are correct. Just as 10^1 is a one with 1 zero after it (or a 10 with zero zeros, if you prefer).

a 10 with zero zeros

dammit! A 10 with zero zeros after it. Big difference.

This won't make any difference to the IDiots. They already "know" everything (as they've declared. And *we're* the pompous assholes? Right.) Our lame, insufficient attempt at the universe is flawed because humans are flawed...and humans do the research = ALL research is flawed = goddidit

We're dealing with a culture that rejects knowledge. They're repulsive.

By Bethistopheles (not verified) on 14 May 2010 #permalink

KingUber said:

Creationists aren't going to be swayed by this, they'll just say that God created every organism similarly.

Indeed -- I have found that creationists will deny that 1+1=2 if they think the Bible says otherwise.

As to addressing the point that common functions required God to use the same code in two different organisms (why? God can do anything he wants, right?), there are a few ways to address that. Many amino acids have more than one codon encoding. Those are somewhat free variables that are useful for showing descent. There are the inactive / corrupt parts of the genome that God had no reason to preserve, yet the variances within these areas show the exact same pattern of inheritance implied by the other evidence for common descent.

As has been said before, if God did create all the creatures, he went to an awful lot of work to make it look exactly as if we all have a common ancestor.

By idiotiddidit#5116d (not verified) on 14 May 2010 #permalink

@KOPD

Thanks, I wasn't sure if I was missing something in the context, or if it had something to do with the statistic (which is definitely not a strong area for me).

I suppose it was put that way because saying "1 in 1 followed by 2680 zeros" would sound kind of weird. Myself, I would have put the explanation for 10^2680 in parenthesis, but it probably is just me being pedantic. What's one zero, more or less? ;)

By matthew.james.neil (not verified) on 14 May 2010 #permalink

Just as a point of comparison, there are approximately 10 to the 78th total atoms in the entire Universe, a puny number compared to these odds.

While I appreciate the thought experiment, the Creationist crowd will ignore it completely, find some middling thing they don't like about it, or say it doesn't matter. Their capacity for cognitive dissonance really is something worthy of serious study.

By https://www.go… (not verified) on 14 May 2010 #permalink

They'll just say God used similar ingredients to make everything and that's why we're similar.

To me the real question has always been: Why aren't we made of animate mud? If god can do anything why'd he use all these tricky bits when we could easily get about with the internal structure of a manikin or a sand castle? And wouldn't that make the existence of God so much more satisfying? Everyone would just assume there's a god because how else could six foot long bags of packing material write symphonies and win the Olympics.

By airbagmoments (not verified) on 14 May 2010 #permalink

Seriously PZ. Bogus mathematics is a lot easier than the other version. You can learn it in seconds, you never have to worry about the consistency or accuracy and it's applicable to just everything. You won't even have to learn how to count... there's no way to beat something like that.

Drat. I'd had some fondness for the hypothesis that bacteria and archaea had independent origins. If life arose independently twice on our planet, the odds that we might find it nearby on other planets would be somewhat higher. Trouble is, it ain't so.

By The Other Ian (not verified) on 14 May 2010 #permalink

Steve L remarked

I like it. But it seems to basically come down to parsimony.

Not exactly. What's being estimated is p(D|H), the probability of the observed data given the various (phylogenetic) models. Hang in for Theobold's forthcoming Panda's Thumb post.

As we know, Occam's Razor is a dull instrument. I think to make inroads, the story will have to be spiced up with virgins and aliens.

Ahhh. p(virgins|model). My bet is that damned few models are virgins. :)

Borel would be disgusted to see his name attached to these moronic creationist blatherings.

By sobaku.no.gaara (not verified) on 14 May 2010 #permalink

Unfortunately, these numbers are entirely consistent with the "deceptive creator" hypothesis. The controversy continues.

By mothwentbad (not verified) on 14 May 2010 #permalink

And you know what we have to do with controversies... Teach them!!

By airbagmoments (not verified) on 14 May 2010 #permalink

The Other Ian, this study doesn’t address the number of times that abiogenesis occurred on Earth. For all we know, abiogenesis, in some simplistic form, could still happen every so often to this day here on Earth.

What it does appear to make emphatic is that all life shares a single common ancestor, much as all humans share a single common ancestor, all primates share a single common ancestor, all mammals share a single common ancestor, and so on.

At the time of the ur-ancestor the study points to, there may well have been many different forms of life from many different instances of abiogenesis. Indeed, I would suggest that it’s highly likely that such was the case, though I have nothing but a hunch to base that suggestion on.

However, our shared great…grandparent is the only one whose descendants survive to this day. I would also suggest that this ultimate patron (and its spawn) ate all the competition, and that we continue to do so to this day.

Cheers,

b&

--
EAC Memographer
BAAWA Knight of Blasphemy
``All but God can prove this sentence true.''

By Ben Goren (not verified) on 14 May 2010 #permalink

::: eagerly awaits the next Luskin "After-Giggle" ™ :::

By darvolution pr… (not verified) on 14 May 2010 #permalink

Just to add to Ben's comment, just because we haven't found any life that doesn't share a common ancestor doesn't mean that it's not out there somewhere. Take for example when they found all the life around the hot vents on the bottom of the ocean. Although the chances have got to be fairly slim.

By Doug Little (not verified) on 14 May 2010 #permalink

our shared great…grandparent is the only one whose descendants survive to this day.

And even that might not be true, as some biologists have suggested that independent lineages might still exist in some extreme niches that are not exploited by conventional life (the shadow biosphere" notion).

idiotiddidit#5116d @ # 28: ... I have found that creationists will deny that 1+1=2 if they think the Bible says otherwise.

Your biblical research has evidently gone deeper than mine. The closest I've gotten is the claim that 1=3 (in some reading-between-the-lines penumbral chapter & verse). As it's well-known that pi=3 biblically, my own calculations show that pi=1, but so far I haven't ... oh wait.

Are you trying to declare that pi+pi=2?!? Heretic! Blasphemer!

By Pierce R. Butler (not verified) on 14 May 2010 #permalink

…and, just to expand further on the point that Doug added to, a lot of it has to do with how you define the term, “life.” Is a bacterium alive? A virus? A prion? A strand of DNA? RNA? Where do you draw the line?

The answer is that life is a lot like porn. You know it when you see it, but good luck actually giving it a meaningful definition. We may well come across some interesting chemistry somewhere that, left to its own for a few billion years, would eventually write blogs. Is it alive? If not, when does life actually begin?

(“When the dog dies and the kids go to college.”)

Cheers,

b&

P.S. In case it isn’t obvious, the answer to that question in the context of the abortion debate can only rationally be answered, not “at conception” or “at birth” or somewhere inbetween, but, “It began once a few billion years ago and hasn’t stopped since.” You may have been born when you emerged from your mother’s womb, but your life — and mine — began about ten billion years after the Big Bang. b&

P.P.S. There’s a non-trivial chance somebody reading this post may live to see many instances of biogenesis, initiated by humans. I refer not just to biological experiments but also to inorganic computer / machine life. Again, “What is life?” b&
--
EAC Memographer
BAAWA Knight of Blasphemy
``All but God can prove this sentence true.''

By Ben Goren (not verified) on 14 May 2010 #permalink

One problem with this approach is that Theobald ignores the mechanistic 'model' that creationists use.

and why shouldn't he?

it's like saying ignoring that someone claims 2+2=grapefruit is a problem for mathematics.

no, Doug didn't do this paper for, or because of, creationists. He did it because of the people arguing about the importance of horizontal gene transfer in models of common descent. I think PZ even made comment to that effect, as did Doug himself over on the thread on PT.

frankly, as much as we laugh at it here, creationism is entirely irrelevant to the actual practice of science.

But they all share those proteins because of god, therefore god exists! Yay! I win the god proofs contest!

By MadScientist (not verified) on 14 May 2010 #permalink

Take for example when they found all the life around the hot vents on the bottom of the ocean.

current evidence suggests these organisms are not independently derived.

current evidence suggests these organisms are not independently derived.

To put it more bluntly, they're ordinary annelids, ordinary crustaceans, ordinary oysters, and ordinary hydrogen-sulfide-eating oxygen-breathing bacteria.

That's another important part: everything in a hot-vent ecosystem breathes oxygen, which comes from photosynthesis in higher layers of the ocean. The claims of total independence were overblown; they only refer to the fact that the production of organic matter from carbon dioxide doesn't involve photosynthesis down there.

By David Marjanović (not verified) on 14 May 2010 #permalink

@#44

Life = rapid, on-going series of chemical reactions which make & break organic molecules?

(Fire: not life because (1)although rapid, products are not organic molecules (unless you count buckminsterfullerines) and (2) not ongoing over long periods of time

Crystals: not life because extremely slow reactions again not producing organic molecules)

Don't know if that definition would precisely fit the RNA hypothesis at life's beginnings ....

Just seems to me it is the fundamental difference between, for example, a live raccoon in the woods and a dead one on the highway, and between something that has never been "alive".

All other definitions seem to be emergent qualities that most groups, but not all individuals, exhibit from that fact, instead of precisely what each living organism which is alive is able to do because of those rapid, ongoing chemical reactions which make and break organic molecules.

What, still no creobots like SP, mendacious delusionalist, to attempt to show their unpublished pull-it-out-of-their-ass calculations are better than the peer reviewed literature? COWARDS!

By Nerd of Redhead, OM (not verified) on 14 May 2010 #permalink

@ #19 - Thx for the Intercessors web-site - entering prayers for God to answer

I saw a good one:

I pray that Jehovah Lord of All will give me the insight on how much to ask when I sell my daughter into slavery Exodus 21 7 and 11! Also before I sell her I pray to the Lord that she does what Lot's daughters did to him, namely get me drunk and sleep with me Gen 19 31 and 32. Amen!

By JustShowMeTheData (not verified) on 14 May 2010 #permalink

AHAHHAHA!

Dear Lord if you only have one flying robot squid don't forget I asked first. They also make some awesome helium filled flying robot penguins so that would be my second choice if there are no squid available

oh my youz guyz are so much fun!

http://www.theintercessors.com/prayers.php?page=7

more more!

By Artemis the Protein (not verified) on 14 May 2010 #permalink

Steve L #12,

(Hey, were Adam and Jesus brothers? Or clones? ... Meh, why start asking questions now -- I'll never get it.)

I never realised how obvious this question should be, but it must be an old one because it's settled dogmatically (and vapidly, in the sense of being 'noncognitive' or 'not even wrong') in the Creed (at least the version I was taught), where Jesus is specified as being 'begotten, not made'. Adam doesn't get a specific mention but Genesis reports he was made (not begotten, there being no available virgins to rape at that time).

By John Scanlon FCD (not verified) on 15 May 2010 #permalink

PZ Myers:

It's a common creationist tactic to fling around big numbers to 'disprove' evolution: for instance, I've had this mysterious Borel's Law (that anything with odds worse than 1 in 1050 can never happen) thrown in my face many times,

Argh! Creationists can't even understand Borel's law of large numbers if that's what they say it means.

Because according to Wikipedia, here's what it really says:

Roughly speaking, Borel's law of large numbers, named after [French mathematician] Émile Borel, states that if an experiment is repeated a large number of times, independently under identical conditions, then the proportion of times that any specified event occurs approximately equals the probability of the event's occurrence on any particular trial; the larger the number of repetitions, the better the approximation.

In other words, the more you repeat the experiment, the closer the actual observed proportion of result E gets to the probability of finding result E.

For example, if you throw a coin 100 times in the air, the proportion of getting "heads" is closer to 0.5 probability than if you throw the same coin 10 times. And it's even closer if you throw it 1000 times. Etc.

By irenedelse (not verified) on 15 May 2010 #permalink

I have it on good reference that the odds of your particular quirky religous belief structure being the One True real one are around one in 750 million, even after graciously assuming that the conditional probability of Planet Earth and Homo sapiens being correct are 1.0.

Theobald's enormous probability that life did not have a common ancestor is misquoted (under understated by 180 orders of magnitude) by both National Geographic and PZ. The odds are 10 to the 2,860th power in favor according to the paper (not 10^(2,680) as reported with the 6 and 8 transposed—a mistaken difference of 10^180, itself much larger than the so-called Borel's Law). Theobald's paper itself:

According to a standard objective Bayesian interpretation of the model selection criteria, the scores are the log odds of the hypotheses. Therefore, UCA is at least 10^(2,860) times more probable than the closest competing hypothesis. Notably, UCA is the most accurate and the most parsimonious hypothesis.

Humorously, Theobald also includes in Table 1 on page 220 the hypothesis that all life except animals and humans have a common ancestor—essentially a test of Genesis 1:20–27. According to the figures reported in Theobald's paper, the odds of animals arising by separately as in Genesis 1:20 are one in 10 to the 5,264th power, and the odds of humans arising separately as in Genesis 1:27 are one in 10 to the 6,105th power.

The numbers all come from Table 1, "Class I hypothesis of single versus multiple ancestries" on page 220, in the column ΔAIC, the natural log-likelihood of the probability minus the total number of parameters in the model. Just convert to base 10 these base e likelihoods comparing the common ancestry "ABE" hypothesis to the other noncommon hypotheses:

"AE+B" hypothesis (most likely uncommon ancestry): 10^(–2,860) = 10^(–6,586/log(10))

"ABE(–M)+M" hypothesis (common ancestry, except animals): 10^(–5,264) = 10^(–12,120/log(10))

"ABE(–H)+H" hypothesis (common ancestry, except humans): 10^(–6,105) = 10^(–14,057/log(10))

I wouldn't put my eggs in the statistical basket here. Just because the likelihood for one event to have happened exceeds the one for the other event it does not say it actually happened that way. Likelihood is an prediction for an event to happen in the future. Any possible permutation of 100 head and tail combination has the same likelihood prior to throwing a coin 100 times and looking at this number makes it very unlikely to happen, even after you just did it.

Horizontal gene transfer is not off the table at all. Thermodynamics predict life to evolve spontaneously in the proper conditions. Our carbon based life has a rich source of combinatorial possibilities and there is NO REASON WHATSOEVER to eliminate the assumption life could have developed several times on planet earth. For one due to our complete lack of understanding abiogenesis and the underlying mechanisms. There are a bunch of scenarios that could explain the catalysis of (later) self sustainable, organized, organic chemistry which subsequently could have been combined in various forms of pre-life and life.

Just some thoughts:

1. sputnik virophages strongly point to an important role of horizontal gene transfer, at least in virions and thus primitive living matter.

2. What happened to the D-Amino Acids ? Either the amino acid introduced in early life (single origin model) were selected upon chirality by cosmological radiation prior to introduction on planet earth via comets, or there was a different selection criteria from a pool of pure L, pure D or mixed life forms we do not understand, yet (multiple origin model or single origin).

3. M. tuberculosis spends almost all its (nasty) life inside the host and acquired at least 19 host genes during evolution, which is direct evidence for horizontal gene transfer.

Yubal says, "Likelihood is an prediction for an event to happen in the future."

Actually, this is not true. A likelihood is merely the product of the probabilities for a set of random events {x} drawn from some distribution P(x). It allows one to select parameters for a model that gives a "best fit". It also plays a critical role in Bayesian probability. And it is related, via the Akaike Information Criterion to the Kullbach-Liebler distance, which measures the departure of a candidate model from the "true" model. As such, it's well suited to selecting among models.

Unfortunately, it is pretty clear that not all events are drawn from the same distribution, some events are probably correlated, and saying what a particular likelihood means under those circumstances is quit a fraught proposition.

By a_ray_in_dilbe… (not verified) on 17 May 2010 #permalink

Hm, OK, since we obviously have a mathematician here:

- How many critical parameters does the model need to have to be appropriate for said statistical analysis?

- Can one select the "best fit" of a model(-system) one prepared him/herself?

- Can the dependence of the "true model" on the selection criteria be eliminated?