Validity of Statistics

[AdvRoboticsE529]

Not that I'm saying statistics does not work in certain cases, however, as the improvement computer technology and robotics in manufactures, uncertainty in industry can be eliminated.

 

Obviously, previously when computing and robotic technology is premature, most do not have the time for 100% confirmation, however, I hope that this could change.

[davidivad]

@studiot

 

i must concede that with large pours of concrete of the same type they take routine measures to document and verify that the concrete is poured correctly.

 

statistics for quality control...

[studiot]

Six bags?

 

Do you have any idea how much 2 and a half kilotonnes is?

[davidivad]

that would be two and a half kilotonnes.

of six bag.

Edited March 31, 2014 by davidivad

[studiot]

Would you like it delivered to the south pole?

 

[davidivad]

yes, but only six bag.

[Prometheus]

When you can predict with certainty the result of a die roll you might have a case.

[AdvRoboticsE529]

"When you can predict with certainty the result of a die roll you might have a case."

 

It's classical mechanics, depending on the variables, not a new branch of maths...

Edited March 31, 2014 by AdvRoboticsE529

[Prometheus]

So you can do it?

[AdvRoboticsE529]

A problem of a circular disk for example, where the application of force is at the line of symmetry on the 2nd dimensional plane of the disk, the point of force applied and its displacement from the centre of mass is 'd1', force applied is 'x', displacement from ground is 'd2' etc.

 

Trigonometric function for one circular spin of the disk, determine time taken and the displacement.

 

Preferably in a vacuum.

 

If variables are given, why not? If the force is not applied at the line of symmetry in the context of a 2nd dimensional plane it would be more difficult as its spinning motion is less perfect.

Edited March 31, 2014 by AdvRoboticsE529

[studiot]

OK so smileys over we can return to rational discussion of the topic.

 

Robowhatsit, there are many uses (and abuses) of statistics.

 

You seem to have embraced the so called 'clockwork universe' in your thinking. If you have not heard of this you should look it up, it was a milestone in scientific development several hundred years ago.

In its day it was a fine concept, but it has been surpassed in the theoretical world.

 

But back to the practical world.

I have indicated where the proper use of statistics improves the quality of production in industry and commercial activity.

 

But it also has been brought into the design process so that now it is fully integrated.

 

You may, or may not, have heard the phrase 'Limit State Design'.

It is the philosophy that underlies modern design codes and actual practices.

It works like this.

 

We can discuss this rationally in an adult fashion if you like or not at all.

 

Take a bridge (design).

 

A bridge has the longest design life of any man made artifact.

 

So what are the loads you are going to design for?

and how are you going to impose them on your design model?

 

You cannot predict what will attempt to drive over that bridge 100 years after your death.

 

What will the bridge be made of?

 

How can you ensure that every last scrap of material will possess at least the strengths you specify?

 

And what of durability and climate and possible ground movements?

 

What happens when the bridge is 100 years old and the materials of construction have deteriorated with age?

 

I will grant you something that no designer has ever possessed or is ever likeley to possess.

 

A perfectly accurate model at the time of design.

 

How are your calculations coming along?

[davidivad]

not to be a pain, but i would pay someone to do a scale model test.

[AdvRoboticsE529]

I do not dispute that in industry statistics can be useful, that doesn't mean that after the successful establishment of the true relationship that it cannot be improved upon.

You cannot predict accurately what will drive over the bridge a century later, if there is a technological revolution similar to that of the industrial revolution, instead of designing a structure of non-existent technology not of the scale of the technology of future transportation system, this will not be preferable or practical, instead, you should demolish or re-modify.

 

Such that building railways before the existence of magnum trains, you don't apply historical technology but propose a new format.

 

So, the application of such is already flawed.

 

You can ensure the properties of the material over specific unit (a form of "concentration" of quality) by artificial synthesis and with control assisted by computing technology and accuracy driven technology. Consider artificially synthesized diamonds, further, you apply maths to result in 100% certainty, diamonds for example possess a quite perfect structure, not that of statistics.

Also, something in which may seem imperfect, we shouldn't be so quickly to think that there is no definite pattern, consider fractal dimensions, instead of applying statistics and encouraging uncertainty, why not look for them?

[davidivad]

a hammer for a nail and a wrench for a bolt.

it requires both to be the most effective design for anything.

[studiot]

 

We can discuss this rationally in an adult fashion if you like or not at all.

 

 

I am sorry you do not wish to learn about things you know naught about.

 

I have wasted enough time on this thread.

[AdvRoboticsE529]

Sorry for my bad grammar, makes it difficult to get my point across.

 

Statistics can be useful, however, it doesn't mean or seem to be the best method for most models.

[Popcorn Sutton]

 

Are you asking this as someone who understands proper statistics and its proper uses?

 

Statistics makes for a better, safer world by providing quality control tools.

 

There not even a sniff of a definition.

 

Agreed. It depends on your interests, really.

[AdvRoboticsE529]

Not really much comment on the validity of statistics, for most part of this thread there's mostly defensive comments for statistics yet does not prepare to take the other side to conclude in a less biased conclusion.

[Bignose]

Statistics can be useful, however, it doesn't mean or seem to be the best method for most models.

I think you had best define "most". I, for one, would like you to explain how to answer the two problems posited in this thread.

 

1) How to estimate the amount of active drug in pills without statistics? Related: how should the FDA decide is a drug is effective or not without statistics?

 

2) How to estimate the composition of concrete when high volumes are being produced every day without statistics?

 

and then I have a few more

 

3) How to estimate the amount of rainfall that will occur 10 days from now without statistics?

 

4) How to estimate the batting average of the Cardinals' starting third baseman this season without statistics?

 

5) How to estimate how many tractors manufactured this year will break down due to a drivetrain failure without statistics?

 

6) How to estimate whether a patient with a certain amount of cancer should choose between chemotherapy, radiation treatment, or invasive surgery without statistics?

 

I look forward to your answer.

Edited March 31, 2014 by Bignose

[studiot]

 

I look forward to your answer.

 

Me too +1

Edited March 31, 2014 by studiot

[AdvRoboticsE529]

@Bignose

You don't estimate, this would eliminate the purpose.

 

I mentioned possibilities of mass-confirmation with the assistance of computing technology and robotics, although likely still premature, it does not exclude this possibility as well as many other methods, further, the fact that statistics is applied in industry does not validate the method, which most of you don't seem to care.

 

By estimation you no longer have 100% certainty, so, if you have the technology to not to do so, you don't.

 

I already gave an example of statistics being akin to throwing an object 'x' amount of times in a controlled lab to result in a needed variable from false average, where mechanics provide a valid picture of the motion of the object to result in the 100% accurate value for the needed variable, different fields obviously requires different respective methods.

 

Most of you keep repeating the same thing.

By eliminating the purpose, that is, the purpose of replacing statistics.

Edited April 1, 2014 by AdvRoboticsE529

[studiot]

Since you eschew statistics and also estimation (which is not the same) what is you opinion of approximation theory?

 

Worse, you are looking at mechanics through rose tinted spectacles.

 

There are many problems for which the equations cannot be solved.

 

Worse still there are problems in mechanics that have no explicit equations to solve.

 

I am not saying that statistics is the answer to these, for it is not, but over to you for your opinion.

Edited April 1, 2014 by studiot

[AdvRoboticsE529]

I have no opinion on the approximation theory, the purpose for the replacement of statistics should be to eliminate uncertainty, hence estimations will not be preferred regardless of their relations.

 

You use too much metaphors, and is very unspecific, difficult to discuss with you, I try. You also shouldn't be so presumptuous in that many problems cannot be solved with "explicit" equations, this is the mentality in which should be discouraged, and is encouraged by statistics, just because certain problems seems difficult currently does not mean it is unsolvable, and should be worked on towards certainty in contrast to living in uncertainty. I previously asked my teacher how could you determine the steepest gradient of any given function, she said it is not possible (I searched online for a method, which is beyond my current skill yet is possible), other questions I have asked includes the summation of root numbers which is premature yet does not mean it is not possible to be precise in the formulation of formulas, you are not unlike many authorities who are confident that what they know is the truth, relativity shows that time is relative, a contrast to the previous authorities who refused such concepts.

 

Such that the proof of Fermat's last theorem seems unsolvable with previous methods and seems only possible with the advent of computation, I would happily assume that all problems are solvable and can be shorten into a simple equation / formula, after the rigorous proofs that makes a very interesting journey.

 

Anyways, a thread originally to see the general opinion of the validity of statistics, not really going where it should.

[John Cuthber]

statistics must be interpreted.

like art, it requires pattern recognition which is subjective.

this is why it is the job of the scientist to be as objective as possible.

in a paper that has statistics, and is well written, the author does not quantify the data for you. he or she merely shows you the data.

it is up to the professionals of the field to decide if it is useful.

since right and wrong are subjective ideas in themselves, we pick the interpretations of those best able to pick the least of evils.

truth be told the concept is effective enough to have measurable results.

That post contains a number of misconceptions- sadly very common ones.

Statistics is - like the rest of maths- objective.

You can apply a number of statistical tests to some data and get different outcomes.

There are two reasons for that

1) some of the tests are simply not appropriate.

2) the tests have different powers and- they always give an "answer" that's probabilistic in nature so they can legitimately differ.

 

It is the job of scientists to be objective.

It's also their job to either know what statistical methods to apply or to get help from a statistician.

The problem is that most scientists think they know what they are doing, so they don't ask the experts unfortunately they also overestimate their own abilities and use the wrong methods and tests..

 

so, when you say "it is up to the professionals of the field to decide if it is useful."

Do you mean you should choose the right tools for the job (in which case I agree) or are you saying

"keep on doing different tests until you get the answer you want" ?

in which case you are talking about the antithesis of science.

 

 

It's also bizarre to claim, on a science website that

"right and wrong are subjective ideas in themselves"

The right answer in science or maths is the right answer.

[Bignose]

By estimation you no longer have 100% certainty, so, if you have the technology to not to do so, you don't.

So you're saying that you'd rather not know the answer at all if you can't know with 100% certainty? I guess I'd much, much rather have estimates for answers to all the questions I asked above rather than just give up because they couldn't be known with perfect accuracy. Way to duck the issue too. Maybe you ought to change your stance from 'most' to 'virtually none'.

 

I didn't even break out the real big question: how do you plan to do quantum mechanics calculations without statistics? There is absolutely no '100% certainty' in doing QM calculations, yet its been pretty darn successful (as have all the other calculations I mentioned above) -- including helping design the electronics you are using to post here.

 

Anyways, a thread originally to see the general opinion of the validity of statistics, not really going where it should.

Really? You can't figure it out from the context? Oh wait, that would require some estimation/inferring/approximating wouldn't it? Well, let me spell it out for you with 100% certainty.

 

My opinion -- and I suspect many others' here -- is that statistics is perfectly valid. And I suspect your opinion of their non-validity is based on some instances where statistics have been misinterpreted, abused, or misunderstood. It is also my opinion that there is a gross naivity on the number of models that allow for perfect calculation as well the number of instances where anything anywhere can be known with 100% certainty.

 

It is really humorous that you keep asking about the non-validity of statistics, when, in fact, the mathematics of statistics themselves can and are known with 100% certainty. The mathematics of statistics start from axioms like all other math and the formulas are derived directly from those axioms with proofs and everything. Math is nice like that, all the statements can be proven. What you really keep asking about is the non-validity of statistics in their application to the real world. A fair question, to be sure, but I think you really need to think about what you're saying before you insist on 100% assurances to answer questions. Think about that the next time you take any medicine, ever. There is no way to be 100% sure that it will be effective on whats infecting you. And, there is no way to be 100% sure that what is infecting you will kill you. But, I for one would rather tips the odds in my favor via medication than waiting for 100% confidence. Feel free to live you life how you want, though.