Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.advocacy > #114958

Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line

Cross-posted to 4 groups.

Show all headers | View raw

On 6/18/12 12:26 PM, in article
2455309a-29d5-441c-93ad-99ac248ced1d@t8g2000yqd.googlegroups.com, "Onion
Knight" <onionknightgot@gmail.com> wrote:

> On Jun 18, 4:32 pm, Steve Carroll <fretwiz...@gmail.com> wrote:
>> On Jun 18, 7:19 am, cc <scatnu...@hotmail.com> wrote:
> 
>> Unquestionably, irrefutably... Snit's "prediction"  that the alleged
>> correlation between rising Linux marketshare and UI improvements were
>> something he labeled as a "trend", another term widely used in
>> statistical analysis. Given the context his numerous statements are
>> being made in (Statistics), a few obvious questions are:
>> 
>> 1 - Did Snit create a statistical model to make his prediction (an
>> alleged correlation) and to claim this alleged "trend"?
> 
> Snit showed there was an increase in 2011 and a decrease in 2012. He
> has admitted that he did not predict the decrease in 2012.

Absolutely correct.

>> 2 - Was Snit aware that the word 'prediction' is a term widely used in
>> statistical analysis, one that has a specific meaning which involves
>> the use of statistical modeling?
> 
> Can you be specific with what you thought his predictions were?

Of course he cannot be.  He is making things up.

My prediction was that there would be an increase in usage - and for the
latter half of 2011 I was right.  For the first half of 2012 I was wrong.

There really was no more specific "prediction".  Carroll just likes to make
things up.

>> 3 - Despite his claim that he took a class on statistics, was Snit
>> even aware that statistical modeling exists?
> 
> You will insist no because you are a liar. Who cares.

Carroll is back to theorizing about my life - he does this to change the
topic.

>> 4 - If 'yes' to point 3, was Snit aware that if a model encompasses
>> the right kind of data it could provide evidence of a "correlation",
>> another term widely used in statistical analysis and one that Snit
>> utilized in his prediction.
> 
> Snit did show the correlation in 2011 with his prediction of an
> increase.

And I have talked about there being all sorts of forms of analysis.
Carroll's admission that he did not know this shows he is ignorant of what
is being discussed - or he is lying.

I go with the latter.

>> 5 - Did Snit believe that evidence of a "correlation" could be shown
>> by using *any* data?
> 
> Whatever he believed he did show the correlation.

Of course I did... though I also showed where the data went against my
statement that Linux usage would increase (while I did not state it that I
recall, the assumption was that it would not then decrease right after...
but I suppose I could play silly games like Carroll and say I never did deny
that... LOL!)

>> 6 - Was Snit aware the the word "trend" (aka trend analysis) has a
>> specific meaning in a discussion involving statistics?
> 
> What trend of his are you speaking of?

The one he made up in his head.  Carroll cannot understand what he reads so
he makes things up and plays word games.

>> One could go on and on but you get the drift...
> 
> Of course we get the drift. You are an obsessive liar who has been
> following Snit around for a decade.

Absolutely correct.  Ever since his then-girlfriend dumped him and he blamed
me.  

>> Notably, the poster 'cc' asked Snit for evidence of his allegation of
>> a correlation and Snit replied with no evidence, just a re-assertion
>> (this is typical for Snit):
>> 
>> "The correlation fits with my prediction." - Snit
> 
> Are you denying that Snit showed an increase of usage in 2011?

Carroll will run from that question.  But I did:

<http://tmp.gallopinginsanity.com/LinuxTrend2011-2ndhalf.png>

>> Snit then asserted that it was cc who didn't know what a "correlation"
>> is, a statement that was proven to be erroneous. The facts show it's
>> Snit who proved he doesn't know what a correlation is, as shown here:
> 
> You are a liar.

Of course Carroll is a liar.   I made it very clear I knew what was a
correlation and that such did not prove causation.  Carroll and his sock
could not understand that (amazing how they both are ignorant in the same
ways).

>> https://groups.google.com/group/comp.os.linux.advocacy/msg/f206d56f7b...
>> 
>> Snit's "prediction", which he *subsequently* claimed was backed by a
>> portion of the data but not by the rest, was built on his pressing
>> need to save face and be right about *something* (anything!). In
>> statistics... how can a "correlation" that has no evidence for its
>> existence fit with a "prediction"? Logically, of course, there is no
>> connection between his 'prediction" and the data, as evidenced by his
>> failure to show the correlation he alleged exists. Don't forget, as
>> shown above, Snit's "prediction *was* that there was a correlation.
>> Period.
> 
> Are you denying that Snit showed an increase of usage in 2011?

Carroll will run from that question.  But I did:

<http://tmp.gallopinginsanity.com/LinuxTrend2011-2ndhalf.png>

>> Snit tried to sprinkle red herring powder all over the discussion
>> throughout numerous threads and posts. He lied when he believed it
>> would help him, he went off topic when he believed it would help him
>> and he used a sock pupptet/shill when he believed it would help him.
> 
> You have accused at least three people of being socks or shills of
> Snit. Can you back any of your lies?

Carroll claims all topics where not even he can find good lies to back his
hatred are "off topic".  But the facts stay the same:

1)  cc was wrong to say I missed steps in the creation of a linear trend
line in Excel.  I did no such thing.

2) cc was wrong to claim the incorrect depictions I showed him of sigma
lines were, in fact, incorrect.  But they were.

3) cc was wrong to say I was pushing the correlations I noted as being proof
of the causation I had spoken of earlier.  I did no such thing.

4) cc was wrong to deny the fact that on a depiction of a normal
distribution you can visually see where the sigma lines should be drawn
based on the distance from the mean (specifically, the distance from the
mean to the inflection points).

>> Snit tried an appeal to authority (notably, he  lied about what the
>> real topic was) and his response to that person was his admission that
>> he had no serious goals for his trend line with respect to the data
>> used to create it.
>> 
>> In others words... business as usual for Snit.
> 
> You follow Snit around lying about him. Yes.  Business as usual.

Exactly correct.

-- 
The indisputable facts about that absurd debate: <http://goo.gl/2337P>
cc being proved wrong about his stats BS: <http://goo.gl/1aYrP>
7 simple questions cc will *never* answer: <http://goo.gl/cNBzu>
cc again pretends to be knowledgeable about things he is clueless about.

Back to comp.os.linux.advocacy | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

Re: Visualizing where to draw the standard deviation line cc <scatnubbs@hotmail.com> - 2012-06-18 06:19 -0700
  Re: Visualizing where to draw the standard deviation line Snit <usenet@gallopinginsanity.com> - 2012-06-18 09:25 -0700
    Re: Visualizing where to draw the standard deviation line Steve Carroll <fretwizzer@gmail.com> - 2012-06-18 09:43 -0700
    Re: Visualizing where to draw the standard deviation line cc <scatnubbs@hotmail.com> - 2012-06-18 09:52 -0700
      Re: Visualizing where to draw the standard deviation line Snit <usenet@gallopinginsanity.com> - 2012-06-18 10:55 -0700
  OT: Summarization of Snit's 'Statistical Analysis' argument  -  Re: Visualizing where to draw the standard deviation line Steve Carroll <fretwizzer@gmail.com> - 2012-06-18 09:32 -0700
    Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Onion Knight <onionknightgot@gmail.com> - 2012-06-18 12:26 -0700
      Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line cc <scatnubbs@hotmail.com> - 2012-06-18 12:42 -0700
        cc runs. Again. Snit <usenet@gallopinginsanity.com> - 2012-06-18 13:11 -0700
          Re: cc runs. Again. cc <scatnubbs@hotmail.com> - 2012-06-18 13:23 -0700
            Re: cc runs. Again. Snit <usenet@gallopinginsanity.com> - 2012-06-18 13:42 -0700
            Re: cc runs. Again. Steve Carroll <fretwizzer@gmail.com> - 2012-06-18 13:56 -0700
              Re: cc runs. Again. cc <scatnubbs@hotmail.com> - 2012-06-18 14:09 -0700
                Re: cc runs. Again. Snit <usenet@gallopinginsanity.com> - 2012-06-18 14:31 -0700
                Re: cc runs. Again. Steve Carroll <fretwizzer@gmail.com> - 2012-06-18 16:04 -0700
      Re: OT: Summarization of Snit's 'Statistical Analysis' argument -  Re:Visualizing where to draw the standard deviation line Frederick Williams <freddywilliams@btinternet.com> - 2012-06-18 20:42 +0100
      Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Snit <usenet@gallopinginsanity.com> - 2012-06-18 13:18 -0700
      Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Steve Carroll <fretwizzer@gmail.com> - 2012-06-18 13:30 -0700
        Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Onion Knight <onionknightgot@gmail.com> - 2012-06-18 13:48 -0700
          Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Snit <usenet@gallopinginsanity.com> - 2012-06-18 14:11 -0700
            Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Steve Carroll <fretwizzer@gmail.com> - 2012-06-18 15:45 -0700
              Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Onion Knight <onionknightgot@gmail.com> - 2012-06-18 21:15 -0700
                Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Snit <usenet@gallopinginsanity.com> - 2012-06-18 23:02 -0700
                Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re: Visualizing where to draw the standard deviation line Steve Carroll <fretwizzer@gmail.com> - 2012-06-19 09:12 -0700

csiph-web