Groups | Search | Server Info | Login | Register

Groups > comp.os.linux.advocacy > #115400

Re: Correct process for creating a linear trend line in Excel

Show all headers | View raw

On 6/20/12 9:50 AM, in article
0192c2ef-cca7-4452-8192-613dce514315@googlegroups.com, "cc"
<scatnubbs@hotmail.com> wrote:

>>> Did you even read the link? The data point doesn't have to be taken in error
>>> to be an outlier.
>> 
>> Yet before you were calling these data points "erroneous"
>> 
>>   cc:
>>     ------
>>     Your process is wrong because you give too much weight to
>>     error signals/erratic data/anomalous data/freak occurances/
>>     whatever term you prefer.
>>     -----
>> 
>> But now you back pedal from that...
> 
> I

You keep running as I respond to your full posts.

No more.  Stop running and *respond* in an honest fashion... otherwise you
are just waving a white flat and that is boring:



On 6/20/12 8:56 AM, in article
6996c5a3-db13-4dab-a0b5-c9ac84ded54b@googlegroups.com, "cc"
<scatnubbs@hotmail.com> wrote:

> On Wednesday, June 20, 2012 11:46:23 AM UTC-4, Snit wrote:
>> On 6/20/12 5:31 AM, in article
>> 0a278b19-7cd1-4231-91b5-0e33462114cc@googlegroups.com, "cc"
>> <scatnubbs@hotmail.com> wrote:
>> 
>>> On Tuesday, June 19, 2012 7:33:59 PM UTC-4, Snit wrote:
>>>> 
>>>> But I also know where I am right... or at least where is *no* evidence of
>>>> my being wrong and plenty of support I am right - such as my prediction of
>>>> an increase in Linux usage and the reported increase in the latter half of
>>>> 2011.
>>>> 
>>> http://pareonline.net/getvn.asp?v=9&n=6
>>> 
>>> Take the average of the data set. Now take the standard deviation. Now find
>>> your outliers. Now realize why that is evidence you are wrong in your new
>>> attempt as well.
>>> 
>> You are trying to excuse your own failure to recognize the trends you
>> missed: Linux usage, based on the data, *did* move upward significantly in
>> the latter half of 2011.  If this has been from a data point that was taken
>> in error then the trend would not be so clear and powerful:
>> 
> 
> Did you even read the link? The data point doesn't have to be taken in error
> to be an outlier.

Yet before you were calling these data points "erroneous"

  cc:
    ------
    Your process is wrong because you give too much weight to
    error signals/erratic data/anomalous data/freak occurances/
    whatever term you prefer.
    -----

But now you back pedal from that... as you run from my responses to you.
Boring.  You just keep changing your story as you try to find *something*,
*anything* that would make it look like you had a clue.

All the while you run from the fact I showed there was a very definite and
strong trend in the latter half of 2011 in terms of Linux usage
increasing... and have also showed where it turned around in 2012.

But all you can do is run:

>> <http://tmp.gallopinginsanity.com/LinuxTrend2011-2ndhalf.png>
>> 
>> Before you were worshipping the god of R^2 values.  Well, that graph has the
>> highest R^2 value of any you or I have generated... and, of course, it uses
>> the same process I have been showing you.  Heck, if I fiddled with the
>> numbers I could make the R^2 value for those 6 months even stronger - just
>> claim that the data that brings it away from having an R^2 value of .999 is
>> just an "anomaly" or that they are outliers.  I *could* play the same games
>> you are.
>> 
>> But I am honest and I shall not do so.  So I show the "simple" linear trend
>> line - the one made as most linear trend lines are made and as is shown by
>> the process provided by MS, Apple, and others.
>> 
>> But your "better" process made you miss this trend - to use your analogy of
>> summer heat - with your focus on looking at the average temperature of a
>> location you missed that there is a trend for summers to be hotter than that
>> average!  Then when this was pointed out to you, you insisted that if someone
>> notes the summer is hotter and shows you a trend line of an increase in
>> temperature for June, July, and August that they *must* be predicting that
>> September, October, and November will continue in the same upward increase in
>> temperatures - because after all that is what trend lines do: they predict
>> the future.  Utter idiocy... but that is what you and my stalker have been
>> arguing.  Just nonsense.
>> 
>> Neither of you had a clue what you were talking about... so you just make
>> things up.  And when others come in and tell you this you cannot handle it so
>> you make up stories about how they are my "socks".  Again, just utter idiocy.
>> You cannot handle when you have been proved wrong.

You just snipped all of the above.  Not even you believe your own BS... if
you did you would not be running so fast.

And below you just spew nonsense to try to inflate your own ego.  Boring.
And dishonest.  My comments have nothing to do with the global warming
deniers.  Nothing at all.  You are just showing off your ignorance.

> The point of taking the average and standard deviation is to
> mathematically identify anomalies, no matter how they occurred. It's
> mathematical evidence that you're basing your trend in the latter half of 2011
> on mostly outliers.
> 
> You're structuring your argument the same way people who try to disprove
> global warming do when they see a cold temperature. The temperature is legit,
> but that's not the problem with trying to use it to disprove global warming.
> 
> I didn't miss any trends, because I know how to handle the extremes.
> Unfortunately you still don't get it. Your original argument has been shot to
> pieces and now you're trying to cling to scraps that don't work either.


-- 
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: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-18 21:33 -0700
  Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-18 23:19 -0700
  Re: Correct process for creating a linear trend line in Excel TomB <tommy.bongaerts@gmail.com> - 2012-06-19 19:07 +0000
    Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-19 12:17 -0700
      Re: Correct process for creating a linear trend line in Excel Steve Carroll <fretwizzer@gmail.com> - 2012-06-19 13:33 -0700
        Re: Correct process for creating a linear trend line in Excel Tattoo Vampire <sitting@this.computer> - 2012-06-19 18:23 -0400
          Re: Correct process for creating a linear trend line in Excel Steve Carroll <fretwizzer@gmail.com> - 2012-06-19 16:13 -0700
          Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-19 16:33 -0700
            Re: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-19 18:32 -0700
              Re: Correct process for creating a linear trend line in Excel Tattoo Vampire <sitting@this.computer> - 2012-06-19 21:44 -0400
                Re: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-19 20:31 -0700
                Re: Correct process for creating a linear trend line in Excel Captain Napalm <fake@email.address> - 2012-06-20 07:20 -0400
                Re: Correct process for creating a linear trend line in Excel chrisv <chrisv@nospam.invalid> - 2012-06-20 06:34 -0500
                Re: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-20 18:28 -0700
            Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-20 05:31 -0700
              Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-20 08:46 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-20 08:56 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-20 09:32 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-20 09:50 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-20 09:57 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-20 10:27 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-20 10:32 -0700
                Re: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-20 18:32 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-20 19:48 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-21 07:42 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-21 09:12 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-21 10:16 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-21 10:55 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-21 11:08 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-21 11:27 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-21 11:58 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-21 13:20 -0700
                cc cannot help but run like a coward. Snit <usenet@gallopinginsanity.com> - 2012-06-21 13:21 -0700
              Re: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-20 18:30 -0700
                Re: Correct process for creating a linear trend line in Excel Snit <usenet@gallopinginsanity.com> - 2012-06-20 19:52 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-21 06:28 -0700
                Three questions for cc Snit <usenet@gallopinginsanity.com> - 2012-06-21 09:21 -0700
                Re: Correct process for creating a linear trend line in Excel cc <scatnubbs@hotmail.com> - 2012-06-21 06:29 -0700
        Re: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-19 18:31 -0700
    Re: Correct process for creating a linear trend line in Excel Onion Knight <onionknightgot@gmail.com> - 2012-06-19 18:29 -0700

csiph-web