Area 51 » Science

218

followers

Announcements

I'd like to propose a change to 'Proposed Q&A site for statistics, data analysis and data mining'. It sounds a bit trivial, but the name as it stand implies that only people who do only statistics for a living should use the site. There are people in many other professions who might have a lot to contribute to the site, but be put off if they feel like it is just for people where statistics is their profession rather than part of their profession. – Ian Turner May 27 '10 at 20:57

Statistics is not a subset of mathematics. It uses some mathematics, but so does physics, engineering, etc. Try asking any of these questions on MO and they will (rightly) be closed as off topic. – Rob Hyndman May 27 '10 at 23:40

I've modified the subheading to take account of the comments being made. Let me know if you think this is better (perhaps vote up this comment). I'm still not sure what to do with the major heading. I understand the objections, but haven't come up with anything better yet. Feel free to make some suggestions in the comments. – Rob Hyndman Jun 1 '10 at 3:55

show 14 more announcements

113 Example Questions (12 closed)

active oldest votes

up vote 0 down vote

added by Jon Seigel May 28 '10 at 0:30

up vote 0 down vote

added by Jon Seigel May 28 '10 at 0:31

up vote 0 down vote

added by Harvey Motulsky May 28 '10 at 3:04

I posted this to start a discussion about analyzing molecular biology data. I think analysis of sequences, homology, genetic linkage, etc. are in another world and belongs on another site. – Harvey Motulsky May 28 '10 at 3:05

... like biostar.stackexchange.com. – Matt Parker May 28 '10 at 6:19

up vote 0 down vote

added by Jay Stevens May 29 '10 at 12:24

up vote 0 down vote

added by Anon Jun 2 '10 at 3:26

up vote 0 down vote

added by Anon Jun 2 '10 at 3:53

closed as off topic by Matt Parker, chollida, Rob Hyndman, Peter K., Mike Mazur Jun 5 '10 at 10:33

This question does not relate to the topic of the proposal.
It was closed as part of an automated migration of off-topic to close votes on September 29, 2011.

	Pretty clearly a question for StackOverflow. When I'm optimizing code, I want computer geniuses picking it over. – Matt Parker Jun 3 '10 at 0:01
	That was the intent! I wanted it be declared as off-topic as I agree with your assessment.To my surprise most of us saw it as on topic. :-) – Anon Jun 3 '10 at 0:21

up vote 0 down vote

added by Menno Jun 2 '10 at 9:10

Probably too general a question. – Pedro J. Aphalo Jun 8 '10 at 13:28

up vote 0 down vote

added by robin girard, edited by WikiSpeedia hang-around Jun 3 '10 at 21:58

up vote 0 down vote

added by Mehper C. Palavuzlar Jun 3 '10 at 10:59

up vote 0 down vote

added by Simon Nickerson Jun 3 '10 at 19:04

closed as off topic by dswan, Shane, Pedro J. Aphalo, Peter Smit, Harvey Motulsky Jun 8 '10 at 16:26

This question does not relate to the topic of the proposal.
It was closed as part of an automated migration of off-topic to close votes on September 29, 2011.

Too variable. it depends on the country, how many licenses you buy and the price at the time you ask. – chollida Jun 3 '10 at 22:55

up vote 0 down vote

added by Peter Smit Jun 8 '10 at 10:30

Maybe one sentence doesn't catch this question completely. Here randomization is meant in the context of statistical significance testing. – Peter Smit Jun 16 '10 at 17:26

up vote -1 down vote

added by Rob Hyndman May 25 '10 at 2:17

closed as off topic by Noah Snyder, Joe Mako, Shane, François G. Dorais, Maxim Zaslavsky Jun 8 '10 at 19:31

This question does not relate to the topic of the proposal.
It was closed as part of an automated migration of off-topic to close votes on September 29, 2011.

A better question might be "What factors determine the unemployment rate? – Troggy May 27 '10 at 4:20

This question sits on the thin line between on- and off-topic. As an econometrician, I would be interested in where I can find unemployment rates for certain countries. So, this could be an on topic question if you replace the "what is" by "where can I find". – Waleed Al-Balooshi May 27 '10 at 12:11

For me it is totally off topic. The site should be about data analysis etc. rather than how to gether the data. Getting general data covers just too many fields that are probably better suited to their own Stack Exchanges. – Ian Turner May 27 '10 at 12:40

I agree with Splash. Questions like "How can I identify factors associated with the unemployment rate?" seem on-topic to me, and maybe "How are unemployment rates calculated?", but this is a reference question. Just because a question refers to a statistic doesn't mean it's a question about statistical analysis. – Matt Parker May 27 '10 at 18:22

This is exactly what Wolfram Alpha is for. Incidentally it's 4.1% (world rank 157th) (2008 estimate) – blissapp Jun 2 '10 at 23:12

up vote -1 down vote

added by Harvey Motulsky May 25 '10 at 14:14

No tech support here. This site should be for statistical principles, not details of one particular program. – Harvey Motulsky May 25 '10 at 14:15

up vote -1 down vote

added by Harvey Motulsky May 25 '10 at 14:16

	This is trickier, but I think this is a slippery slope to a boring site. I'd keep it focussed on questions about analyses, not tools. – Harvey Motulsky May 25 '10 at 14:17
	Like you say, if you start forbidden EVERYTHING that's not 'theoretical' you might end up with a very boring site. I would leave it be – Ivo Flipse May 26 '10 at 21:32
	I think general tool questions are ok. I wouldn't want to see tech support questions for a specific tool, but I think it is useful to allow questions on what tool to use for a specific type of analysis. – Rob Hyndman May 27 '10 at 6:29

up vote -1 down vote

added by Mike Mazur May 27 '10 at 2:54

	For example: how likely is it that Steve will be in his office at 10:00am based on his presence history on IRC? – Mike Mazur May 27 '10 at 2:56
	I voted no. The idea of this site is to ask questions that have answers, not questions that require a text book to answer. Not to discuss perspectives. This one seems way too vague. If someone asked it, others might be able to point him towards the right text book. But there is no way to answer a question like that, so it doesn't really fit. One instance of that question like that is: What is the chance that the global mean temperature will be more than 4 degrees C warmer than now in 50 years? Many books have been written on that. – Harvey Motulsky May 27 '10 at 3:04
	I'm not a statistician, so please take this with a grain of salt. I think this question has merits. One potential answer could be: Try a markov model. Here's a link. Or, use Bayseian statistics, etc. It could end up being a very useful resource on the web. – Scott Wisniewski May 27 '10 at 3:39
	Voting no because it is too vague. A more specific question with context might be ok. – Rob Hyndman May 27 '10 at 3:44
	I agree Scott, it just needs to be more specific. I'll add a similar question that I think would be more appropriate. – Rob Hyndman May 27 '10 at 4:32

up vote -1 down vote

added by Juan Antonio May 27 '10 at 5:30

up vote -1 down vote

added by Ian Turner May 27 '10 at 11:52

closed as off topic by george, Peter K., Rob Hyndman, Jeromy Anglim, Anon Jun 5 '10 at 12:45

This question does not relate to the topic of the proposal.
It was closed as part of an automated migration of off-topic to close votes on September 29, 2011.

Sounds like homework – Rob Hyndman May 27 '10 at 11:57

I actually took this question from here: newscientist.com/article/…. It is actually more involved than you think, but really is a statisitics 'game'. I asked the question to see if users want this sort of game on the Stack Exchange. – Ian Turner May 27 '10 at 12:04

This site is about data analysis, not probability. But a few brain teasers will make it more fun. More fun = more visitor = more chance of critical mass. So I vote yes. – Harvey Motulsky May 27 '10 at 13:32

Wait, isn't this one of those problems where people just can't seem to agree on the outcome? (Like the Monty Hall Problem) – Menno May 27 '10 at 18:01

Probability goes hand-in-hand with statistics, so I don't see why that subject would be off-topic. That said, this is a bad question. – Shane Jun 7 '10 at 13:42

up vote -1 down vote

added by cletus May 27 '10 at 23:29

up vote -1 down vote

added by Jay Stevens May 29 '10 at 12:26

up vote -1 down vote

added by Stu Thompson May 30 '10 at 9:08

	(Or 'how can i do something dodgy in the world of statistics analysis') – Stu Thompson May 30 '10 at 9:09
	Do you mean STATA? – Rob Hyndman Jun 2 '10 at 2:01
	More approroate for STATA tech support or SO perhaps? – Anon Jun 2 '10 at 3:07
	STATA, yes. (Been working with physical SATA hard drives lately :P) The edit feature is broken for me again...will write Area51 kids. – Stu Thompson Jun 2 '10 at 5:17
	@Anon: Write STATA about how to hack their own software? Did you read the questions? – Stu Thompson Jun 2 '10 at 5:17

show 3 more comments

up vote -1 down vote

added by Scott Wales Jun 2 '10 at 1:44

up vote -1 down vote

added by csgillespie Jun 2 '10 at 9:44

It would be fine, I think, if the purpose of the comparison and type of model would have been specific. This question is too general. – Pedro J. Aphalo Jun 8 '10 at 11:49

up vote -1 down vote

added by robin girard Jun 2 '10 at 11:18

up vote -1 down vote

added by Mehper C. Palavuzlar Jun 3 '10 at 10:35

too broad perhaps? what application do you want to use them for? Do you need to have the numbers be reproducible? Do you want one that steps? What languages are you willing to use? Are you willing to pay for software? – chollida Jun 3 '10 at 19:33

Sounds like a valid question to me; especially for someone not familiar with the options. – Mal Ross Jun 3 '10 at 19:40

Most statistical packages come with high-quality random number generators, The days of poor PRNGs bolloxing up simulations went out with RANDU. That issue is virtually a historic accident. The question can be answered by 'anything that uses Mersenne Twister', which (these days anyway) means 'Just about everything'. – ConcernedOfTunbridgeWells Jun 10 '10 at 11:24

up vote -1 down vote

added by robin girard, edited by Rob Hyndman Jun 3 '10 at 13:25

up vote -1 down vote

added by Pedro J. Aphalo Jun 8 '10 at 13:42

up vote -1 down vote

added by Michel de Ruiter Jun 15 '10 at 20:10

up vote -1 down vote

added by Michel de Ruiter Jun 15 '10 at 20:13

up vote -1 down vote

added by robin girard Jun 16 '10 at 5:58

Too vague; without knowing more about what a "power forecast" is, it's impossible to answer. This might be more appropriate in a solar power area. – pinko Jun 16 '10 at 21:04

up vote -2 down vote

added by Menno May 27 '10 at 9:06

Not sure about this one... on the one hand this kind of falls under 'Statistics 101', on the other hand it could turn into a nice resource for people who want to ask that kind of question (and thus no longer have to) – Menno May 27 '10 at 10:15

I think it is definitely the type of resource that should end up on this Stack Exchange, but the current system does not suit this sort of question very well. Community wiki's might help. I would also avoid not allowing questions that are classed as '101'. Some people have a lot of perception of a problem without necesserily having much theoretical understanding and I think we should encourage that sort of person as well as giving them the tools to improve their theortical understanding. – Ian Turner May 27 '10 at 13:34

advanced statistics - to vague – Karsten W. Jun 6 '10 at 15:00