From: William Hughes on 14 Feb 2010 19:11 On Feb 14, 6:30 pm, JSH <jst...(a)gmail.com> wrote: > On Feb 14, 10:26 am, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > On Feb 14, 12:28 pm, JSH <jst...(a)gmail.com> wrote: > > > > On Feb 14, 4:48 am, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > On Feb 14, 1:10 am, JSH <jst...(a)gmail.com> wrote: > > > > > > On Feb 13, 10:41 am, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > > On Feb 13, 1:09 am, JSH <jst...(a)gmail.com> wrote: > > > > > > > > On Feb 12, 8:21 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > > > > On Feb 12, 10:27 pm, JSH <jst...(a)gmail.com> wrote: > > > > > > > > > > On Feb 12, 5:59 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > > > > > > On Feb 12, 8:46 pm, JSH <jst...(a)gmail.com> wrote: > > > > > > > > > > > > On Feb 11, 7:08 pm, William Hughes <wpihug...(a)hotmail..com> wrote: > > > > > > > > > > > > > JSH: Well if the predictions are wrong then > > > > > > > > > > > > JSH: there is no further argument. End of story. > > > > > > > > > > > > > Are your predictions wrong? Please start your answer Yes or No. > > > > > > > > > > > > Statistical arguments bore me. ... > > > > > > > > > > > Try again. > > > > > > > > > > No. > > > > > > > > > Ok. Thats clear enough. You think that saying > > > > > > > > 1.12 is 1 is not wrong. > > > > > > > > Please elaborate for the physics people. What exactly do you mean > > > > > > > with those numbers? > > > > > > > To anyone who can draw a distinction between "not wrong" > > > > > > and "wrong but maybe close enough to be useful" > > > > > > (this group includes physics people) you are saying 1.12 > > > > > > is 1. > > > > > > In probability is anything 1? > > > > > > My point is your posts indicate you believe that 100% correctness is a > > > > > reasonable goal for an event that is about probability, so you're > > > > > saying that a 112% result is "wrong". Is that correct? Begin with > > > > > yes, or no please. > > > > > Yes. For instance to say that the probability that a fair coin > > > > shows heads is .49 is wrong. > > > > Are you sure about that? > > > Yes. Just as sure as I am that saying > > "4 is prime" is wrong. A fair coin is defined to > > have a probability of .5 of showing heads. > > > - William Hughes > > So what if you flip it a thousand times and it ends up 0.49? > > Is the coin not fair? If it IS fair, is 0.5 "wrong" as the > probability? Should it instead be 0.49? > > Which is "right"? Probability is not defined by the result from one thousand trials. "The probability is 0.5" is right. The prediction for the number of heads in one trial of one thousand flips is wrong. (No surprise, the prediction will almost always be wrong). -William Hughes
From: JSH on 15 Feb 2010 00:37 On Feb 14, 4:11 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > On Feb 14, 6:30 pm, JSH <jst...(a)gmail.com> wrote: > > > > > > > On Feb 14, 10:26 am, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > On Feb 14, 12:28 pm, JSH <jst...(a)gmail.com> wrote: > > > > > On Feb 14, 4:48 am, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > On Feb 14, 1:10 am, JSH <jst...(a)gmail.com> wrote: > > > > > > > On Feb 13, 10:41 am, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > > > On Feb 13, 1:09 am, JSH <jst...(a)gmail.com> wrote: > > > > > > > > > On Feb 12, 8:21 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > > > > > On Feb 12, 10:27 pm, JSH <jst...(a)gmail.com> wrote: > > > > > > > > > > > On Feb 12, 5:59 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > > > > > > > On Feb 12, 8:46 pm, JSH <jst...(a)gmail.com> wrote: > > > > > > > > > > > > > On Feb 11, 7:08 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > > > > > > > > > > > JSH: Well if the predictions are wrong then > > > > > > > > > > > > > JSH: there is no further argument. End of story. > > > > > > > > > > > > > > Are your predictions wrong? Please start your answer Yes or No. > > > > > > > > > > > > > Statistical arguments bore me. ... > > > > > > > > > > > > Try again. > > > > > > > > > > > No. > > > > > > > > > > Ok. Thats clear enough. You think that saying > > > > > > > > > 1.12 is 1 is not wrong. > > > > > > > > > Please elaborate for the physics people. What exactly do you mean > > > > > > > > with those numbers? > > > > > > > > To anyone who can draw a distinction between "not wrong" > > > > > > > and "wrong but maybe close enough to be useful" > > > > > > > (this group includes physics people) you are saying 1.12 > > > > > > > is 1. > > > > > > > In probability is anything 1? > > > > > > > My point is your posts indicate you believe that 100% correctness is a > > > > > > reasonable goal for an event that is about probability, so you're > > > > > > saying that a 112% result is "wrong". Is that correct? Begin with > > > > > > yes, or no please. > > > > > > Yes. For instance to say that the probability that a fair coin > > > > > shows heads is .49 is wrong. > > > > > Are you sure about that? > > > > Yes. Just as sure as I am that saying > > > "4 is prime" is wrong. A fair coin is defined to > > > have a probability of .5 of showing heads. > > > > - William Hughes > > > So what if you flip it a thousand times and it ends up 0.49? > > > Is the coin not fair? If it IS fair, is 0.5 "wrong" as the > > probability? Should it instead be 0.49? > > > Which is "right"? > > Probability is not defined by the result from > one thousand trials. "The probability is 0.5" is right. Why? What makes it "right"? > The prediction for the number of heads > in one trial of one thousand > flips is wrong. (No surprise, the prediction will > almost always be wrong). > > -William Hughes Then what good is this probability thing then? If it's "almost always" "wrong"? What IS probability? What is "prediction"? James Harris
From: William Hughes on 15 Feb 2010 07:35 On Feb 15, 1:37 am, JSH <jst...(a)gmail.com> wrote: > On Feb 14, 4:11 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > <snip> > > The prediction for the number of heads > > in one trial of one thousand > > flips is wrong. (No surprise, the prediction will > > almost always be wrong). > > Then what good is this probability thing then? If it's "almost > always" "wrong"? If probability is used to predict the exact number of successes in N trials the prediction will almost always be wrong. However, there are other types of predictions (e.g. the number of successes will be within k of E) which are almost always right. - William Hughes
From: Junoexpress on 17 Feb 2010 01:45 On Feb 15, 12:37 am, JSH <jst...(a)gmail.com> wrote: > On Feb 14, 4:11 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > > > The prediction for the number of heads > > in one trial of one thousand > > flips is wrong. (No surprise, the prediction will > > almost always be wrong). > > > -William Hughes > > Then what good is this probability thing then? If it's "almost > always" "wrong"? > > What IS probability? What is "prediction"? > > James Harris The probability of interest refers to the probability that a sample mean varies from the population mean by a given amount. This is a very well-posed and also well-understood (to everyone except you of course) problem. If you're confused as to where to start, go back and see if you can first grasp the concept of a "confidence interval". After you understand this little pearl of wisdom, you should then try to understand the different types of convergence for an estimator. Then maybe you can understand a nice little result called the "Weak Law of Large Numbers". Then hopefully your brain fog will clear up a bit and then maybe, just maybe, you might have some non-trivial thoughts to share with us on probability (although the odds of that are probably pretty slim). HTH, M
From: JSH on 17 Feb 2010 01:54
On Feb 16, 10:45 pm, Junoexpress <mtbrenne...(a)gmail.com> wrote: > On Feb 15, 12:37 am, JSH <jst...(a)gmail.com> wrote: > > > On Feb 14, 4:11 pm, William Hughes <wpihug...(a)hotmail.com> wrote: > > > > The prediction for the number of heads > > > in one trial of one thousand > > > flips is wrong. (No surprise, the prediction will > > > almost always be wrong). > > > > -William Hughes > > > Then what good is this probability thing then? If it's "almost > > always" "wrong"? > > > What IS probability? What is "prediction"? > > > James Harris > > The probability of interest refers to the probability that a sample > mean varies from the population mean by a given amount. This is a very > well-posed and also well-understood (to everyone except you of course) <deleted> Arguments are tedious. The real problem here is that you are not a scientist, and neither is William Hughes. Have your fun with stupid arguments on Usenet. Science is something you can just read about, and imagine you understand. Scientists pose questions and look for answers. If you wish, I can concede the argument to you like I tried to do with William Hughes earlier. Ok? You win. Isn't that what you wanted? Argument over. James Harris |