T'ai Chi : Some math/stat questions for you

Re: Re: Re: T'ai Chi : Some math/stat questions for you

T'ai Chi said:
That's almost correct. Student was the pseudonym of a William Sealy Gosset, a statistician who worked analyzing samples at the Guiness brewing factory. He did his work with the t-distribution to allow for analyzing small samples- something to do with yeast if I am remembering correctly. Guiness didn't want their competitors to know that they had a huge advantage by having a statistician working for them, hence the pseudonym.

This information is found in the very first link in the results returned by google when you search on the term "student-t distribution". Pretty much the same wording, too, except rearranged a bit.
 
T'ai Chi said:


In my opinion, I think this question is a little more vague than the rest. I can give you some general guidelines here, if you'd like. (even though biostatistics is not really my area)



Hey, if someone calls me stupid, I will stick up for myself, and then call them on it. [/B]


How about this:

The null hypothesis is that there are no people with supernatural powers.

The alternate hypothesis is that, although rare, there are people with supernatural powers.

How would you test this?

An alternative:

If we assume that there are people with supernatural powers, design a study to determine the prevalence of these powers.

I hope that's better.
 
Re: Re: Re: Re: T'ai Chi : Some math/stat questions for you

roger said:

This information is found in the very first link in the results returned by google when you search on the term "student-t distribution". Pretty much the same wording, too, except rearranged a bit.

Um... just how distinctively can you phrase such elementary knowledge?

Don't let your beliefs about me get in the way of the facts.

The Gosset question was quite easy to answer. I think that the information is in the majority of the statistics books I own, and it is also information that is brought up a lot in lectures.

I've even brought it up in some of the lectures I've given on the subject.
 
TruthSeeker said:

How about this:

The null hypothesis is that there are no people with supernatural powers.

The alternate hypothesis is that, although rare, there are people with supernatural powers.

How would you test this?


Well, that is obviosuly very vague.

It might be better for you to explain what you mean by "supernatural powers" and then choose one of those "supernatural powers" specifically to test.

Then we could talk about specific designs, etc.
 
Re: Re: Re: T'ai Chi : Some math/stat questions for you

T'ai Chi said:
In any case, if Z is a standard normal random variable, then Z^2 is a chi-square random variable with 1 degree of freedom.
That's just a book definition. Surely you can explain in simple english terms the relationship.
 
1. Let f(x) be differentiable on [a,b]. At x=0, the function's concavity changes. Is it a relative minimum or an inflection point? What test do we apply here?

This was the only question I was very close to understanding, But I didn't even quite understand it.

1. Does on [a,b] mean in the interval between a & b?

2. If the concavity changes is that by definition an inflection point?

3. If the concavity changes does that preclude the possibility that it's a relative minimum?

4. Does the reference to the first derivative test mean to find out if the first derivative at the point in question is equal to zero in which case we will be either at a local minimum or a local maximum?

I am hoping that at some point each of the other questions will be discussed in a little more detail, but I didn't want to ask too many questions because I didn't want to get in the way of a pissing contest.

edited to add: albeit one that seems kind of entertaining.
 
Originally posted by davefoc
1. Does on [a,b] mean in the interval between a & b?
Yes. More precisely, it's the closed interval from a to b. That means it includes the endpoints a and b. The notation is standard. The notation for open intervals, i.e., those that don't include their endpoints, uses parentheses instead of square brackets, like this: (a, b). You can also use one square bracket and one parenthesis, e.g., [a, b) or (a, b], to denote a half-open interval, i.e., an interval that includes one endpoint and excludes the other.
2. If the concavity changes is that by definition an inflection point?
Yes, at least according to the two calculus books I just checked.
3. If the concavity changes does that preclude the possibility that it's a relative minimum?
I'm pretty sure that's correct, unless f doesn't have a first derivative at x = 0, in which case a relative minimum is definitely possible. The question allows that case, as it never states that 0 is in [a, b], though the omission was probably an oversight.
4. Does the reference to the first derivative test mean to find out if the first derivative at the point in question is equal to zero in which case we will be either at a local minimum or a local maximum?
I don't understand the reference to the first derivative test. For that matter, I don't understand the question, "What test do we apply here?". My first reaction was, "Well, that depends. What do you want to know?"

A zero first derivative at a point doesn't mean the point is a local maximum or minimum. However, at any local maximum or minimum, the derivative will be zero if it exists at all, which it needn't.

Mathworld has a page about the first derivative test.
 
davefoc said:

, but I didn't want to ask too many questions because I didn't want to get in the way of a pissing contest.

edited to add: albeit one that seems kind of entertaining.

Who knew math/statistics could be this exciting. :)

Anyway davefoc, feel free to interject etc. Bill nor I own this post. :)
 
Re: Re: Re: Re: T'ai Chi : Some math/stat questions for you

Iconoclast said:

That's just a book definition. Surely you can explain in simple english terms the relationship.

Let the goalpost moving process begin.

I answer the questions completely and correctly, then you ask for it in "simple english".

:rolleyes:
 
TruthSeeker said:
While I think I know what you are after in Question 3 (we'll see if I'm correct), it did make me laugh.

I'd like to add one question: How big a nerd do you have to be to laugh at stats questions on a Sunday morning?

Sign me up for the old nerd patrol, then.

Fit me to a 'T', man.
 
Re: Re: Re: Re: Re: T'ai Chi : Some math/stat questions for you

T'ai Chi said:


Let the goalpost moving process begin.

I answer the questions completely and correctly, then you ask for it in "simple english".

:rolleyes:

So what would you use, sindscal, indscal, or portscal?
 
Re: Re: Re: Re: Re: Re: T'ai Chi : Some math/stat questions for you

jj said:

So what would you use, sindscal, indscal, or portscal?

I must be an idioscal to reply to you.

Anyway, I don't do anything in Fortran, only SAS, S+, and a few others.
 
Re: Re: Re: Re: Re: T'ai Chi : Some math/stat questions for you

T'ai Chi said:


Let the goalpost moving process begin.

I answer the questions completely and correctly, then you ask for it in "simple english".
Not moving the goalposts. If you truly understand how Chi Squared Distributions are related to Normal Distributions, then you will be able to explain it in simple terms.
 
davefoc wrote:

quote:
--------------------------------------------------------------------------------
1. Let f(x) be differentiable on [a,b]. At x=0, the function's concavity changes. Is it a relative minimum or an inflection point? What test do we apply here?
--------------------------------------------------------------------------------



This was the only question I was very close to understanding, But I didn't even quite understand it.

1. Does on [a,b] mean in the interval between a & b?

2. If the concavity changes is that by definition an inflection point?

3. If the concavity changes does that preclude the possibility that it's a relative minimum?

4. Does the reference to the first derivative test mean to find out if the first derivative at the point in question is equal to zero in which case we will be either at a local minimum or a local maximum?

69dodge did a nice job answering these questions, but to expand on what he said:

4) The first derivative equalling zero will indicate a relative minimum, a relative maximum, or an inflection point ( which is technically both a maximum and a minimum ).

As an example, let f(x)=x^2
then f'(x) = 2x
and f''(x) = 2

At x=0 f'(x)=0, indicating a max, min, or inflection point at x=0.
For all values of x, f''(x)=2, indicating positive concavity and a local minimum.

Then let f(x)=x^3
then f'(x)=3x^2
and f''(x)=6x

Again, at x=0, f'(x)=0, indicating a max, min, or inflection point at x=0.
However, f''(x) varies with x, and f''(x)=0. This indicates a change in concavity, and you check this by finding the values for f''(x) on either side of the point where f'(x)=0. Where x<0, f''(x)<0, indicating negative concavity and a relative maximum. For x>0, f''(x)>0, indicating positive concavity and a relative minimum. This means that for f(x)=x^3, that x=0 is an inflection point.

I hope this helps,

Eric
 
Re: Re: T'ai Chi : Some math/stat questions for you

T'ai Chi said:

1. 1st derivative test[/b]
Very good.

2. It is the opposite. We approximate the binomial distribution with a normal distribution.
Oops. You missed the question. Bone up on CLM.

3. The t-tools are used in many situations. -Mostly when a measure of variance is unknown/estimated, or the sample sizes are small. They are used all the time by professionals and by students, especially in comparing the means of two groups, and in testing regression coefficients. Also, knowledge of the t-distribution is important because the t-distribution is related to or makes up other distributions that are used.
You seem to still be ripping/reading and posting here, as with your brewery post that was very similar to things found on the net. It gets worse here, though. The t-test can be used in the situtation you described, but it is far from the usual use. The most usual use is to compare two samples for sameness.
4. Ratios of the standard deviations (their squares, the variances anyway) are used to determine the relative efficiency of experiments.
We're not talking about the F-test here. I asked about efficiency. Cramer-Rao.
 
For those of you wondering, somebody PMd me to tip me off that T'ai blew his cover in another thread and signed one post "-Who". After that, he wrote that he PMd hal, asking that his "T'ai Chi" sock puppet be deleted.

But, like I said. The mask came off long ago.
 
BillHoyt said:
But, like I said. The mask came off long ago.
This isn't really important, but I'm curious: was Sherlock Holmes another of Whodini's sockpuppets?
 
BillHoyt said:
For those of you wondering, somebody PMd me to tip me off that T'ai blew his cover in another thread and signed one post "-Who". After that, he wrote that he PMd hal, asking that his "T'ai Chi" sock puppet be deleted.

But, like I said. The mask came off long ago.
Which thread contains this "confession"?
 

Back
Top Bottom