T'ai Chi : Some math/stat questions for you
I think it's pretty obvious that T'ai Chi does know his stuff. Trying to catch him out on minor details or wording is pretty desperate stuff.
Also I would point out that it's not a very accurate test either. If you were to quiz me like this I would perform pretty miserably off the top of my head. From this you would conclude without a shadow of a doubt that I do not have a stats qualification. And you would be wrong as I have a degree in Mathematics with Statistics from the University of Southampton. However I wasn't the hardest working student and everything I did know at the time has since become very rusty!
Personally I am impressed that T'ai Chi still remembers all this stuff. Unless you are involved with it every day in your career? That would help to keep it fresh!
Also I would point out that it's not a very accurate test either. If you were to quiz me like this I would perform pretty miserably off the top of my head. From this you would conclude without a shadow of a doubt that I do not have a stats qualification. And you would be wrong as I have a degree in Mathematics with Statistics from the University of Southampton. However I wasn't the hardest working student and everything I did know at the time has since become very rusty!
Personally I am impressed that T'ai Chi still remembers all this stuff. Unless you are involved with it every day in your career? That would help to keep it fresh!
UKBoy,UKBoy1977 said:
I don't think you saw how miserably T'ai did. His answer to Number 1 was the only correct answer he gave. His Gosset answer was so close to a rip and re-post from some web sites that it was laughable.
Now the funniest answers were garnered by jj's question. I guess everybody missed the back-and-forth between jj and I. T'ai (Who) certainly did. "Eigen stand it", "vanishing" were all clues that neither "Fortran" nor "English" were correct answers. The first two (indscal, sindscal)) are math packages that have been written in various programming languages. The third one (portscal) was jj's plant:
Um, but portscal doesn't exist.
If T'ai is going to pretend to knowledge, he needs to get better at it.
Now I don't personally care that T'ai doesn't know certain things. I do care that he get a basic premise of skepticism: authority doesn't matter. His howls of protest that I (and others) refuse to cite our degrees are incredbily arrogant as well as utterly specious. The degrees don't matter, T'ai.
I am a bouncer at a local strip club. That's my story and I'm sticking to it. Now, T'ai, deal with the facts, with science, with the truth. Stop worrying about who can scan in fake degrees, and who plays doctor out here on the internet.
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: T'ai Chi : Some math/stat questions f
You want me to apologize for being tricked?
Do get real, fella.
T'ai Chi said:
Gees, JJ.
I replied a post of yours from above a bit:
(bolded by me)
Goodness, gracious me, you didn't say what you're quoting. Why, why you TRICKED me! (gasp)
Oh, poor me.
(Yes, it may have been accidental on your part, but given your performance and dishonest use of sock puppets elsewhere, you'll have to forgive me for being skeptical here.)
Sheesh.
So "let us note" when you will apologize for your error.
You want me to apologize for being tricked?
Do get real, fella.
BillHoyt said:
Awww, you told him. I was waiting to see what else he came up with.
And, no, "English" isn't the right answer, either.
Btw, indscal and sindscal (and a few others ending in 'scal' are all old multidimensional scaling packages written at Bell Labs in the 1970's. They are old and cranky (and some of them are written originally in Fortrash, but have been since transcribed), but at least until recently were some of the more useful ways for extracting data from difference judgements.
I have no idea what's become of Doug and Sandy, I know Sandy retired quite cheerfully a while ago (can't blame her either), so I doubt I can say more about where this stuff is now. Maybe Tez could find it.
(Which is a shame, but like many old packages, it's been tossed because it isn't GUI, doesn't look fancy, etc, but has better math built into it by the people who figured it out the first time. Ditto for the filter design package that Jim Kaiser wrote. The user interface was from h**l, but the math behind it was better than anything I've used to present, including the matlab packages, to say the least.)
Brian the Snail
Muse
- Joined
- Sep 14, 2002
- Messages
- 962
jj said:
Sorry jj, but I don't quite get this. Are you saying that these packages are obsolete?
UK, don't expect them to end. 
Any questions I do answer, Bill can always say that I copy and pasted it, or got it from a book, or asked someone else, etc. There is always some excuse they have.
Perhaps you should have asked me a more difficult question then, Bill, I dunno.
Information about Gosset is common knowledge to statisticians and even statistics students. I TA'd a couple of lower division statistics classes over the last couple of years and that information was common knowledge in them.
Next time I'll complexify my answer for you. I'll type it in Pinyin, Latin, or portscal or something.
Well congradulations to jj, I guess, for stumping me. To think that all people who majored in mathematics/statistics are familiar with all known algorithms is pretty amusing. How many algorithms do you think the common statistics package uses? Personally, I've done very little in multidimensional scaling myself. I've heard of indscal from somewhere, in tinkering with Fortran, or perhaps as some updated function in S+??. I'm not too shocked that I didn't know about that. It wasn't really even a stat question, in my opinion.
Why do you keep insist I am pretending and faking this, Bill? So jj got me on a algorithm question that isn't common knowledge (even among statisticians), um, oookkkk.
Yeah, you'll just start a thread about me though, with the purpose of specifically trying to stump me on math/stat questions.
Oh, arrogant, oh, that hurts.
And suggesting to me that I burn my degrees and spank my professors is not arrogant? Pot? Kettle? Espresso maker? Press? Bill's callin'.
If you flat out tell me to burn my degrees, spank my professors, etc., I will certainly, and justifiably, tell you where to stick it. I have proof of my education and knowledge. The degrees certainly matter, as they show I've invested time, money, effort into higher learning and have graduated from two distinguished universities and mastered that knowledged, obtained a scholarship and a full assistantship, and several awards, and a decent GPA.
And fetched donuts for professors.
Great. I have degrees in mathematics and statistics. Bouncers are capable of knowing statistics, I have no problem acknowledging that. How come I don't question your claim of being a bouncer, but you insist on questioning mine, Bill? What's up with that.
Hey, I would argue with jj about the big physics questions. But for me to question jj about the basic physics of electricity (since he has, in addition, an MSEE and massive amounts of real-world experience) is just askin' for a spankin' that I don't want.
(edited to change my error, "Ph.D" to "MSEE")
I've been dealing with facts, science, and the truth all this time.
You can claim victory over my alleged copy and tweaking of the Gosset information (your irrational speculation), and over jj's reference to portscal (something which isn't basic knowledge, even among statisticians) if you'd like. I'd find that pretty funny.
Any questions I do answer, Bill can always say that I copy and pasted it, or got it from a book, or asked someone else, etc. There is always some excuse they have.Perhaps you should have asked me a more difficult question then, Bill, I dunno.
Information about Gosset is common knowledge to statisticians and even statistics students. I TA'd a couple of lower division statistics classes over the last couple of years and that information was common knowledge in them.
Next time I'll complexify my answer for you. I'll type it in Pinyin, Latin, or portscal or something.
Well congradulations to jj, I guess, for stumping me.
To think that all people who majored in mathematics/statistics are familiar with all known algorithms is pretty amusing. How many algorithms do you think the common statistics package uses? Personally, I've done very little in multidimensional scaling myself. I've heard of indscal from somewhere, in tinkering with Fortran, or perhaps as some updated function in S+??. I'm not too shocked that I didn't know about that. It wasn't really even a stat question, in my opinion.Why do you keep insist I am pretending and faking this, Bill? So jj got me on a algorithm question that isn't common knowledge (even among statisticians), um, oookkkk.
Yeah, you'll just start a thread about me though, with the purpose of specifically trying to stump me on math/stat questions.
Oh, arrogant, oh, that hurts.
And suggesting to me that I burn my degrees and spank my professors is not arrogant? Pot? Kettle? Espresso maker? Press? Bill's callin'.If you flat out tell me to burn my degrees, spank my professors, etc., I will certainly, and justifiably, tell you where to stick it. I have proof of my education and knowledge. The degrees certainly matter, as they show I've invested time, money, effort into higher learning and have graduated from two distinguished universities and mastered that knowledged, obtained a scholarship and a full assistantship, and several awards, and a decent GPA.
And fetched donuts for professors.
Great. I have degrees in mathematics and statistics. Bouncers are capable of knowing statistics, I have no problem acknowledging that. How come I don't question your claim of being a bouncer, but you insist on questioning mine, Bill? What's up with that.
Hey, I would argue with jj about the big physics questions.
But for me to question jj about the basic physics of electricity (since he has, in addition, an MSEE and massive amounts of real-world experience) is just askin' for a spankin' that I don't want.(edited to change my error, "Ph.D" to "MSEE")
I've been dealing with facts, science, and the truth all this time.
You can claim victory over my alleged copy and tweaking of the Gosset information (your irrational speculation), and over jj's reference to portscal (something which isn't basic knowledge, even among statisticians) if you'd like. I'd find that pretty funny.
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: T'ai Chi : Some math/stat que
Hey, it was worth a shot.
No, I applaud you for being skeptical. I don't applaud you for temporarily not realizing that people can reply to posts that are up the page a bit, and then claiming that I somehow tricked you.
jj said:
Hey, it was worth a shot.
No, I applaud you for being skeptical. I don't applaud you for temporarily not realizing that people can reply to posts that are up the page a bit, and then claiming that I somehow tricked you.
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: T'ai Chi : Some math/stat que
Well, I try to stay in context. Some others seem to find that, shall we say, restrictive? Or is the term "confining"?
T'ai Chi said:
No, I applaud you for being skeptical. I don't applaud you for temporarily not realizing that people can reply to posts that are up the page a bit, and then claiming that I somehow tricked you.
Well, I try to stay in context. Some others seem to find that, shall we say, restrictive? Or is the term "confining"?
Brian the Snail said:
Mostly they're lost, and although they are old, no, they are NOT particularly obsolete...
The problem is that they are lost. They are the kind of thing that you need once a year, and sooner or later they disappear.
T'ai Chi said:Next time I'll complexify my answer for you. I'll type it in Pinyin, Latin, or portscal or something.
Well, there's always that good old standby, "anibus fidemus".
Personally, I've done very little in multidimensional scaling myself. I've heard of indscal from somewhere, in tinkering with Fortran, or perhaps as some updated function in S+??. I'm not too shocked that I didn't know about that. It wasn't really even a stat question, in my opinion.
It's a good historical question...
Hey, I would argue with jj about the big physics questions.
Just so's its all clear here. I don't have a PhD in anything. Yeah, I do have a lot of real-world experience and an MSEE, but if you look down at my "www" button you'll find I've been doing something besides electricity for a while...
Here are some statistics questions (36) that I thought up for you Bill. I tried to include a good mix of hard and easy- randomly arranged, of course.
1. For a mound-shaped distribution, what is a decent way to estimate the standard deviation from the range?
2. If you were to do a test to see if Passing and Failing an examination were independent of Before and After applying a drug, what is the name of a test you would use?
3. If you wanted to see if a digital blood pressure cuff's results can be used in place of a traditional blood pressure cuff's results (ie. are "the same"), what statistic/test would you use?
4. Is it possible to have a two-sided alternative hypothesis for a test using the chi-square distribution?
5. Is it possible for degrees of freedom to be a non-integer?
6. P(x) = e^(a+bx)/[1+e^(a+bx)] is a linear model, true or false.
7. If two individual 2x2 tables in a chi-square test for independence showed non-significance, but the combined 2x2 table (ie, literally table 1 + table 2) showed significance, how would you interpret that?
8. Why is testing if the population correlation is 0 equivalent to testing if the population slope coefficient is 0?
9. If we only observe the outcomes from coin flipping (Heads = 1, Tails = 0): 1000100010101, what is the most sensible estimate of the probability of Heads? What general mathematical technique would you use here?
10. If I am studying bugs in a statistics class, what statistics am I studying?
11. What was the name of the person who gave an example of the correlation coefficients be the same for multiple sets of data, but their plots looking completely different?
12. If X's are distributed normally, what do you do to them to get to a log-normal distribution?
13. If X and Y are individually distributed as chi-squares, what do you do to them to get to a F distribution?
14. If X and Y are individually distributed normally, what do you do to them to get to Cauchy?
15. How do you transform between a similarity measure and a distance measure?
16. If X's are distributed as an Exponential, what do you do to them to get to a Double Exponential?
17. Are partial derivatives important in asymptotic normality? If so, how?
18. What are the conditions for a distribution belonging to the Regular Exponential Family?
19. How do you determine if a statistic, T(x), is sufficient for estimating, say, the parameter, theta?
20. Why is ancillarity important in theoretical statistics?
21. Canberra and Bhattacharyya formulas are used for...?
22. What is a sensible plotting technique in a repeated measure analysis?
23. Why is the F distribution called an F distribution?
24. Let's say forty subjects are randomly assigned to four treatment groups, ten to each group. Three responses are measured on each subject. What specific distribution would you use to draw inferences about the differences between means in the different treatment groups?
25. Who worked with the quincunx?
26. Does this make sense: "The probability of the population mean being in the interval [26.4, 28.8]mg's is 95%"?
27. What statistician had a ladder?
28. What are the differences between a confidence interval, prediciton interval, and a statistical tolerance interval?
29. Is a MVUE accurate, or precise?
30. For what distributions is range/SD >= sqrt(2) ?
31. How do you interpret: "The Pearson correlation coefficient of age and gender is .96."
32. What is the average of a dataset with n elements if you repeatedly take a 1st level Winsorized mean?
33. Why do we worry about S (the sample standard deviation)? Why not just focus on S^2?
34. If X_i = K*Y_i+C, then what is the mean and the standard deviation of X?
("_i" is a subscript)
35. What is the general name for the mean of the means?
36. Show mathematically that for a standard normal random variable Z, that the variance of Z is 1.
1. For a mound-shaped distribution, what is a decent way to estimate the standard deviation from the range?
2. If you were to do a test to see if Passing and Failing an examination were independent of Before and After applying a drug, what is the name of a test you would use?
3. If you wanted to see if a digital blood pressure cuff's results can be used in place of a traditional blood pressure cuff's results (ie. are "the same"), what statistic/test would you use?
4. Is it possible to have a two-sided alternative hypothesis for a test using the chi-square distribution?
5. Is it possible for degrees of freedom to be a non-integer?
6. P(x) = e^(a+bx)/[1+e^(a+bx)] is a linear model, true or false.
7. If two individual 2x2 tables in a chi-square test for independence showed non-significance, but the combined 2x2 table (ie, literally table 1 + table 2) showed significance, how would you interpret that?
8. Why is testing if the population correlation is 0 equivalent to testing if the population slope coefficient is 0?
9. If we only observe the outcomes from coin flipping (Heads = 1, Tails = 0): 1000100010101, what is the most sensible estimate of the probability of Heads? What general mathematical technique would you use here?
10. If I am studying bugs in a statistics class, what statistics am I studying?
11. What was the name of the person who gave an example of the correlation coefficients be the same for multiple sets of data, but their plots looking completely different?
12. If X's are distributed normally, what do you do to them to get to a log-normal distribution?
13. If X and Y are individually distributed as chi-squares, what do you do to them to get to a F distribution?
14. If X and Y are individually distributed normally, what do you do to them to get to Cauchy?
15. How do you transform between a similarity measure and a distance measure?
16. If X's are distributed as an Exponential, what do you do to them to get to a Double Exponential?
17. Are partial derivatives important in asymptotic normality? If so, how?
18. What are the conditions for a distribution belonging to the Regular Exponential Family?
19. How do you determine if a statistic, T(x), is sufficient for estimating, say, the parameter, theta?
20. Why is ancillarity important in theoretical statistics?
21. Canberra and Bhattacharyya formulas are used for...?
22. What is a sensible plotting technique in a repeated measure analysis?
23. Why is the F distribution called an F distribution?
24. Let's say forty subjects are randomly assigned to four treatment groups, ten to each group. Three responses are measured on each subject. What specific distribution would you use to draw inferences about the differences between means in the different treatment groups?
25. Who worked with the quincunx?
26. Does this make sense: "The probability of the population mean being in the interval [26.4, 28.8]mg's is 95%"?
27. What statistician had a ladder?
28. What are the differences between a confidence interval, prediciton interval, and a statistical tolerance interval?
29. Is a MVUE accurate, or precise?
30. For what distributions is range/SD >= sqrt(2) ?
31. How do you interpret: "The Pearson correlation coefficient of age and gender is .96."
32. What is the average of a dataset with n elements if you repeatedly take a 1st level Winsorized mean?
33. Why do we worry about S (the sample standard deviation)? Why not just focus on S^2?
34. If X_i = K*Y_i+C, then what is the mean and the standard deviation of X?
("_i" is a subscript)
35. What is the general name for the mean of the means?
36. Show mathematically that for a standard normal random variable Z, that the variance of Z is 1.
If I pulled a coin out of my pocket, flipped it a bunch of times, and got that exact sequence of heads and tails, I would estimate, quite sensibly in my opinion, that the probability of heads is very nearly 1/2.
What would your estimate be?
What would the general mathematical technique tell you, if you flipped the coin once and it came up heads? Is that a sensible answer?
Number Six
JREF Kid
- Joined
- Sep 5, 2001
- Messages
- 5,016
69dodge said:
Yes, that question kinda straddles math and philosophy. I mean, if it were "You saw a dichotomous event and you say this sequence of yes and no" then you'd have one estimate of the probability of the event. But if the "event" is coin flipping then it's a matter of how much you assume...that is, can we assume the coin is fair (and if so, by definition it's 50/50) or can we not assume anything (in which case, get the estimate straight from the data).
Very strictly speaking I think the correct thing to do would be the latter since we weren't told to assume anything about the coin.
davefoc
Philosopher
Number six,
I think you're right. The point of the question is to assume that the coin flip is not the standard 50/50 proposition and to assess what information is available from the given sequence. I was looking forward to the answer.
observation:
This pissing contest is more entertaining than most. I am looking forward to the responses to T'ai chi's questions.
I think you're right. The point of the question is to assume that the coin flip is not the standard 50/50 proposition and to assess what information is available from the given sequence. I was looking forward to the answer.
observation:
This pissing contest is more entertaining than most. I am looking forward to the responses to T'ai chi's questions.