Continuation Part 20: Amanda Knox/Raffaele Sollecito

[Vixen]

It reminds me of the scene on Chicago where Lucy Liu comes home to find her lover Harry in bed with 2 women and the guy says "he's alone" she replies "Alone? You got 2 other women in bed with you." Harry says, "c'mon doll, you going to believe what you see or what I tell you?"

If Vixen can't look at those images and see for herself what's true, she's as deluded as Harry hopes Lucy is.

Lucy was smarter. She trusted her own eyes and shot all 3 of them.

Vinci, in the pay of the defense, photoshopped the images, in a blatant attempt to bend evidence. At least Professors Boemia and Rinaldi were objective forensic scientist simply doing their job.

What is it with the kids defense, they had to cheat?

 

[Vixen]

Game theory is not a subset of statistics. Game theory mostly involves deterministic games with rational players. Thus checkers would be played by game theorists and roulette by irrational statisticians. Game theory mostly involves set theory, vector analysis, graphing theory and geometry/topology; not statistics. That you think game theory is just statistics is what makes me think I have a better grasp of statistics than you.

Sollecito's DNA on the bra was low copy number. (Estimated as less than 200pg although Steffanoni never properly quantified the DNA.)

How do the external third party witnesses tell us what was going on in the flat? How do you resolve the contradiction of Popovic, Curatalo, and Kokomani seeing Knox and Sollecito outside of the flat for most of the evening with the requirement for them to be undertaking a prolonged period of harassment, assault, staging etc. inside the flat. How do you resolve the contradiction of the occupants of the broken down car and recovery vehicle hearing and seeing no activity?

Oh pur-leese. Your grasp on statistics is so poor you think it's all about the normal distribution. Game theory is actually just another arm of probability theory. There is more to it than Gaussian equations. Depending on how one constructs the prisoner dilemma, say, to fit the Kercher case, where we have all three of the alleged perps claiming innocence, you might best use a Bayesian a priori probability scenario. Insert your values either in a matrix or a decision tree, or any diagram you like. Your claim "game theory has nothing to do with statistics" is bosh. Anything to do with probability theory is statistics.

It's a pity you cannot see any contradictions in the PIP's version of events. This tells me that any explanation as to why eye witnesses are important, for all their fallibilities, will only fall on deaf ears, as usual.

 

[acbytesla]

Vinci, in the pay of the defense, photoshopped the images, in a blatant attempt to bend evidence. At least Professors Boemia and Rinaldi were objective forensic scientist simply doing their job.

What is it with the kids defense, they had to cheat?

Total bull. But that's not surprising.

 

[Vixen]

According to Vixen the prosecution is just taking a rest.

...They think it's all over...

 

[Methos]

Vinci, in the pay of the defense, photoshopped the images, in a blatant attempt to bend evidence. At least Professors Boemia and Rinaldi were objective forensic scientist simply doing their job.

What is it with the kids defense, they had to cheat?

Evidence for the highlighted part, please...

When did they become "professors"?
Rinaldi is a "dottore" (well, everyone in Italy seems to be one ) and Boemia is adresed as "Isp. Capo" in the nomination.
It's interesting that they were nominated, hired and paid as "consultants for the prosecution", and not called as witnesses in their function as police officers...
So I'd say, that Rinaldi and Boemia weren't "objective forensic scientist simply doing their job.", but hired guns for the prosecution.
And as both parties "hired guns", the question is, who is more believable...
Easily answered... (hint: it's not Rinaldi with his size 37 shoeprint)

 

[Bill Williams]

Oh pur-leese. Your grasp on statistics is so poor you think it's all about the normal distribution. Game theory is actually just another arm of probability theory. There is more to it than Gaussian equations. Depending on how one constructs the prisoner dilemma, say, to fit the Kercher case, where we have all three of the alleged perps claiming innocence, you might best use a Bayesian a priori probability scenario. Insert your values either in a matrix or a decision tree, or any diagram you like. Your claim "game theory has nothing to do with statistics" is bosh. Anything to do with probability theory is statistics.

It's a pity you cannot see any contradictions in the PIP's version of events. This tells me that any explanation as to why eye witnesses are important, for all their fallibilities, will only fall on deaf ears, as usual.

Just chucking in factoids does not help your cause.

A basic google-search shows you are wrong on this one point. Game theory is a discipline all to itself, there are some aspects of game theory which can be analyzed using probability theory.
One is not an "arm" of the other.

http://levine.sscnet.ucla.edu/general/whatis.htm

All this to hold on to your specious claim that there were people reacting to a flurry of activity at the cottage, when the absolute best you can say is that they saw AK and RS at the square, away from the cottage - and if the timing of any of these phantom witnessings is true, that puts AK and RS at the square nearly all evening - in other words, an alibi for not being involved in the flurry of activity at the cottage!

 

[Grinder]

Evidence for the highlighted part, please...

When did they become "professors"?
Rinaldi is a "dottore" (well, everyone in Italy seems to be one ) and Boemia is adresed as "Isp. Capo" in the nomination.
It's interesting that they were nominated, hired and paid as "consultants for the prosecution", and not called as witnesses in their function as police officers...
So I'd say, that Rinaldi and Boemia weren't "objective forensic scientist simply doing their job.", but hired guns for the prosecution.
And as both parties "hired guns", the question is, who is more believable...
Easily answered... (hint: it's not Rinaldi with his size 37 shoeprint)

yes as doctors they have a junior college education plus one year.

in 8 years no one has come up with a shoe that matches the print, except Vixen who has put forward the theory the shoes were home made.

I'm sure the PGP have tried to find the shoes as part of a photo-shop project or perhaps one of their projects for Meredith

 

[Planigale]

Oh pur-leese. Your grasp on statistics is so poor you think it's all about the normal distribution. Game theory is actually just another arm of probability theory. There is more to it than Gaussian equations. Depending on how one constructs the prisoner dilemma, say, to fit the Kercher case, where we have all three of the alleged perps claiming innocence, you might best use a Bayesian a priori probability scenario. Insert your values either in a matrix or a decision tree, or any diagram you like. Your claim "game theory has nothing to do with statistics" is bosh. Anything to do with probability theory is statistics.

It's a pity you cannot see any contradictions in the PIP's version of events. This tells me that any explanation as to why eye witnesses are important, for all their fallibilities, will only fall on deaf ears, as usual.

I give up. The term you want is 'Bayesian prior' not a priori (which just shows you don't actually do Bayesian stats)

Remember the onus is on the prosecution to make a case for guilt BARD, the defence have to make no case at all, just disprove the prosecution case. It is not the responsibility of the judge (as seemed to happen here) to make up an alternative scenario to that put by the prosecution after the defence and prosecution cases have been put - especially since this means the defence have had no opportunity to respond to the scenario.

 

[Bill Williams]

Vinci, in the pay of the defense, photoshopped the images, in a blatant attempt to bend evidence. At least Professors Boemia and Rinaldi were objective forensic scientist simply doing their job.

What is it with the kids defense, they had to cheat?

Photoshpped evidence!? And the judges and public minister charged them with crimes - assuming falsifying evidence is a crime in Italy.

Or could it be that you simply blurt out the first thing that comes into your head? Like promoting Boemia to "professor?

Other than these concerns I'm sure you'll next accuse us of racism for pointing this out!

 

[Vixen]

Total bull. But that's not surprising.

Prof Vinci is from Uni Bari, which tells you all you need to know.

 

[Vixen]

Photoshpped evidence!? And the judges and public minister charged them with crimes - assuming falsifying evidence is a crime in Italy.

Or could it be that you simply blurt out the first thing that comes into your head? Like promoting Boemia to "professor?

Other than these concerns I'm sure you'll next accuse us of racism for pointing this out!

Wrong person. You do know Prof Vinci used a size 42.5 shoe (Raff size) and shoe-horned it (no pun intended) into a 44.5 Rudy size 11.

[excerpt]FV:
I wanted to make a clarification because I examined the ... the model shoe samples out two break we used for comparison, I have to correct because in fact these are size 44 and a half, not 43 and a half as I said, the specimen is here so ... so much so that it is still the sole inked for the tests we have done.
GCM:
Those used for ...
FV:
These ... yes I had erroneously said 43 and a half ...
GCM:
The consulting comparisons.
LM:
He had spoken of 43 and a half during the deposition, corrects the clerical error is 44 and a half.
FV:
Yes it's a clerical error.
GCM:
Please Prosecution for cross-examination.
MC:
Yes about these shoes at this point while we're at the American size?
FV:
Yes, then the American cuts because there is no English Size ...
MC:
The English Cut?
FV:
It is not the same thing Dr.
MC:
It should be 'there all the law.
FV:
No I the meantime I say that the English size is 42.5 then there are ... there is also 9:30.
MC:
As?
FV:
9.5.
MC:
9.5 that would be?
FV:
UK United Kingdom and then in England.
MC:
In England.
FV:
28.5 cm so the measurement expressed in centimeters and then European measure 44.5.
MC:
Yes and you know what was the size UK shoe size or rather the measure covered shoe box found at the home of Rudy Guede and with which then were made of the benchmarking of shoe prints?
FV:
I think it was written 10.
MC:
E '11.
FV:
It 'Excuse 11, 11. 11.
MC:
So why he has used them on 9:30?
FV:
Because this model of shoes was discontinued so already to find it was not easy, we found this and we used that.
MC:
He also knows what measurement is the difference in millimeters, centimeters now I do not know ...
LM:
The 10:30, I read 10:30.
FV:
Between 9:30 and 11 should be a centimeter.

http://themurderofmeredithkercher.com/Francesco_Vinci's_Testimony

(Public domain)

 

[Vixen]

I give up. The term you want is 'Bayesian prior' not a priori (which just shows you don't actually do Bayesian stats)

Remember the onus is on the prosecution to make a case for guilt BARD, the defence have to make no case at all, just disprove the prosecution case. It is not the responsibility of the judge (as seemed to happen here) to make up an alternative scenario to that put by the prosecution after the defence and prosecution cases have been put - especially since this means the defence have had no opportunity to respond to the scenario.

"a priori" is correct.

Courts don't really use probability. In civil courts the balance is 49-51 to win your case. BARD is a much higher standard, but yet again, it is really a case of whose version the judge/s prefer to believe, the prosecution or the defense.

It is as mundane as that.

 

[bagels]

Vixen I found a transcript of Knox's arrest at her rock throwing party you might be interested:

Seattle Police: Mam we got some reports of loud music and rock throwing
Knox: It was a black guy he did it
Seattle Police: Can you describe him
Knox: Sorry I have amnesia, that's my best truth
Seattle Police: And what's your name?
Knox: Foxy Knoxy

 

[LondonJohn]

Nonsense. It's ALWAYS called a Bayesian Prior. NEVER "a priori". Never.

Trust me - my ex is a Nobel laureate in Bayesian statistics. I double-checked the game theory issue with my grandfather John Nash (I used amazing beyond-the-grave communication techniques), who developed game theory to Nobel prize-winning level, and he said without any hesitation "It's not probability theory (although it uses elements of probability theory in constructing the matrices)". I myself am a lowly Emeritus professor of applied mathematics. So please stop arguing.


(For external observers, game theory would be a useful tool for examining many different scenarios in the Knox/Sollecito case and the Kercher murder in general. It could, for instance, be used to examine the rational choices made by Guede on the night of the murder, and to analyse whether or not the behaviour/actions of Guede, Knox and Sollecito in the aftermath of the murder were compatible with the rational decisions of innocent parties, or whether they were only (or more likely) compatible with the rational decisions of guilty parties. The tricky part would be assigning reasonable quantified probabilities and payoffs to each action, but it should be possible. I may have a go at it later on in the year.)

 

[acbytesla]

Nonsense. It's ALWAYS called a Bayesian Prior. NEVER "a priori". Never.

Trust me - my ex is a Nobel laureate in Bayesian statistics. I double-checked the game theory issue with my grandfather John Nash (I used amazing beyond-the-grave communication techniques), who developed game theory to Nobel prize-winning level, and he said without any hesitation "It's not probability theory (although it uses elements of probability theory in constructing the matrices)". I myself am a lowly Emeritus professor of applied mathematics. So please stop arguing.


(For external observers, game theory would be a useful tool for examining many different scenarios in the Knox/Sollecito case and the Kercher murder in general. It could, for instance, be used to examine the rational choices made by Guede on the night of the murder, and to analyse whether or not the behaviour/actions of Guede, Knox and Sollecito in the aftermath of the murder were compatible with the rational decisions of innocent parties, or whether they were only (or more likely) compatible with the rational decisions of guilty parties. The tricky part would be assigning reasonable quantified probabilities and payoffs to each action, but it should be possible. I may have a go at it later on in the year.)

Your grandfather is the American mathematician John Nash? I'm really curious. Did one of his children marry a Limey?

 

[Samzilla]

[qimg]http://i.imgur.com/cpycqQa.jpg[/qimg]

Vixen please quote this image and tell me they were made by different shoes so I can witness the depths of your self delusion.

Posting the window with bars as "proof" there were no bars wasn't enough?

 

[Bill Williams]

Nonsense. It's ALWAYS called a Bayesian Prior. NEVER "a priori". Never.

A quick google check on "Bayesian priori" seems to rule this one in Vixen's favour.

There is such a thing as "Bayesian priori data", and I don't want to do any more than that, because last time I took statistics, Archimedes needed the bathtub.

But it seems both terms - Bayesian prior and Bayesian priori - are in use.

That is all I am doing with this. Point Vixen.

 

[Zephyrgeist]

John Nash? The 'Beautiful Mind' who invented Hex? (Yeah, I know, it was invented simultaneously by two people.)

But still. John Nash? Cool people. Even if he was NUTS. But still...

 

[Planigale]

A quick google check on "Bayesian priori" seems to rule this one in Vixen's favour.

There is such a thing as "Bayesian priori data", and I don't want to do any more than that, because last time I took statistics, Archimedes needed the bathtub.

But it seems both terms - Bayesian prior and Bayesian priori - are in use.

That is all I am doing with this. Point Vixen.

One may talk about 'a priori probability', or Bayesian prior probability usually shortened to Bayesian prior(s). Usually a priori probability is mentioned in the introduction to Bayesian theory. In practice anyone actually doing Bayesian statistics talks about priors. My point being if Vixen actually did Bayesian stats e.g. in R then Vixen would have used the phrase Bayesian prior. Using the phrase a priori suggests that vixen learned frequentist statistics with an intro to Bayes. Bayesian statistics are very useful in social sciences (Vixen did psychology and would probably have learnt about this in relation to social psychology or educational psychology), where mixed effects models for hierarchical data are used.

It is a bit like the observation that a non native speaker might use more grammatically correct forms than a native speaker. Someone actually doing Bayesian stats would never use the phrase 'Bayesian a priori probability' not that the phrase is incorrect rather it is too correct. Of course there is the possibility that Vixen is an anally retentive pedant who has a predilection for using latin phrases but also does bayesian statistics - I wonder what to set the prior at for that?

FWIW on my book shelf (in my bedroom);
An R and S-plus Companion to multivariate analysis
Bayesian models for categorical data
Hierarchical modelling and analysis for spatial data
Data analysis using regression and multilevel/hierarchical models
Applied longitudinal data analysis
Bayesian data analysis
Bayesian logical Data Analysis for the physical sciences
Wavelet methods in statistics with R
bayesian computation with R
model based geostatistics
Bayesian networks and decision graphs
Statistical analysis of environment space time processes
Model-based geostatistics
Bayesian probability theory: Applications in the physical sciences

So I claim the prize for the geekiest bookshelf, and being truly sad that these are by my bed.

PS I am not a statistician.

 

[LondonJohn]

One may talk about 'a priori probability', or Bayesian prior probability usually shortened to Bayesian prior(s). Usually a priori probability is mentioned in the introduction to Bayesian theory. In practice anyone actually doing Bayesian statistics talks about priors. My point being if Vixen actually did Bayesian stats e.g. in R then Vixen would have used the phrase Bayesian prior. Using the phrase a priori suggests that vixen learned frequentist statistics with an intro to Bayes. Bayesian statistics are very useful in social sciences (Vixen did psychology and would probably have learnt about this in relation to social psychology or educational psychology), where mixed effects models for hierarchical data are used.

It is a bit like the observation that a non native speaker might use more grammatically correct forms than a native speaker. Someone actually doing Bayesian stats would never use the phrase 'Bayesian a priori probability' not that the phrase is incorrect rather it is too correct. Of course there is the possibility that Vixen is an anally retentive pedant who has a predilection for using latin phrases but also does bayesian statistics - I wonder what to set the prior at for that?

FWIW on my book shelf (in my bedroom);
An R and S-plus Companion to multivariate analysis
Bayesian models for categorical data
Hierarchical modelling and analysis for spatial data
Data analysis using regression and multilevel/hierarchical models
Applied longitudinal data analysis
Bayesian data analysis
Bayesian logical Data Analysis for the physical sciences
Wavelet methods in statistics with R
bayesian computation with R
model based geostatistics
Bayesian networks and decision graphs
Statistical analysis of environment space time processes
Model-based geostatistics
Bayesian probability theory: Applications in the physical sciences

So I claim the prize for the geekiest bookshelf, and being truly sad that these are by my bed.

PS I am not a statistician.

Exactly. It's not that "a priori probability" is wrong per se - it's that one always refers to a Bayesian prior, and not a "Bayesian a priori (probability)".

It's a bit like saying that Novak Djokovic won the All England Lawn Tennis Club Single Handed Championship of the World in 2015. Nobody would say that, even though it's technically "correct". Rather, they'd say that he won Wimbledon