If I don't find a data set for my f'n paper soon, I'm going to cry...

[bicentennial_baby]

I need to rework my research paper for Econometrics, and I cannot find a data set with: 1) 30 or more observations, 2) 2 continuous and 1 dummy variable, minimum, that is even slightly interesting. It's not even that, it's that the ones I do find have, like, 5000+ observations, 5 continuous and 6 dummy variables...Way more complicated than what I need. x(

So far there is one candidate, and it's mind numbingly boring...Regional differences in teacher salaries and per pupil spending. :puke:

Sigh.

[Xipe Totec]

1. Check out WEKA

http://www.cs.waikato.ac.nz/ml/weka/

And look under datasets.

[bicentennial_baby]

3. Thank you, I will check that out!

:yourock:

[KitchenWitch]

2. Yikes...

Your mind numbingly boring candidate is what we are living here in our school district. I live in a city of about 200,000 people. Of course, the city I live in is about 30 miles from Los Angeles whose population is around 4.2 million. Our school district is considered...get this...RURAL. Therefore the per student funding from the state is only about 60% of the per student funding in the Los Angeles school district.

:eyes:

[bicentennial_baby]

4. I hope I didn't offend!

I meant mind numbingly boring as in, writing a 10 page paper on this topic might put me into a coma. :P

And that is fucked up! x(

[KitchenWitch]

6. Not offended in the least.

I cannot imagine finding 10 pages worth of content using that particular topic, datasets or no.

[bicentennial_baby]

16. It would be mostly statistical analysis...

And yeah... :boring:

I have SO many research papers going simultaneously. It's insane. :crazy:

[Xipe Totec]

5. This one might be interesting

Data Set Information:

Extraction was done by Barry Becker from the 1994 Census database. A set of reasonably clean records was extracted using the following conditions: ((AAGE>16) && (AGI>100) && (AFNLWGT>1)&& (HRSWK>0))

Prediction task is to determine whether a person makes over 50K a year.





http://archive.ics.uci.edu/ml/datasets/Adult

[bicentennial_baby]

7. Oooh, very interesting!

Thank you!! :D

[Whoa_Nelly]

8. NUmbingly boring and hugely influential

I look to those statistics in thinking of future living areas for me.

I may move back to Tucson, a cut of $10k over my current CA salary schedule. But, the other factor is the average cost of living there, and so it makes it quite possibly a choice that can work,even though I would probably have to spend even more out of pocket per student on a monthly basis to supplement education and classroom needs.

Maybe include in that data set the number of teachers entering/leaving per region despite student population growth...?

[flvegan]

9. There are a ton of dummy variables in GD: P.

:P

[bicentennial_baby]

10. !

:rofl:

Too true.... :D

[KitchenWitch]

11. GD:P is the embodiment of the null hypothesis

[bicentennial_baby]

13. !

Ho: Supporters of Candidate X = 0
H1: Supporters of Candidate X RAWK!

:rofl:

[KitchenWitch]

15. *snerk*

:spray:

[bicentennial_baby]

17. I can't believe we're making GDP Stats jokes...

:rofl:

..and not the Gross Domestic Product kind. Maybe this is what my Prof meant when she said we'd understand Stats jokes later in the semester :rofl:

[KitchenWitch]

18. Does this make us geeks, nerds, or dweebs?

:rofl:

[bicentennial_baby]

19. I'll take option A, please

:rofl:

[mike_c]

12. can you do a subset of a more complicated data set...?

Take one of the more complex data sets you already have and excerpt only one potential question for analysis?

[bicentennial_baby]

14. That may be the direction I go...I have plenty to work with in that respect...

I'm going to come back to it in the morning with fresh eyes. Thanks! :)

[leftyclimber]

20. I agree with mike_c.

Take something you have, figure out a question, snag the appropriate variables, plug, and chug. Instant paper (other than the writeup part, of course).

I was looking around to see if I still had my Master's thesis dataset at home for you to borrow, but it appears that I've deleted it. Hopefully I still have a copy on my office computer. Oops.

[bicentennial_baby]

21. If you do, let me know!

:D

Yeah, this is a good idea. I have a ton of datasets that I can pare down in Excel before loading into STATA. Thanks! :)

[leftyclimber]

22. If you're ever interested in doing econometric analysis on recreation stuff

I have access to a bunch of good "loaner" datasets, some of which are moderately sized and some of which are honkin' big. I did a little bit of econometric stuff in a class a few years ago -- it works really well for contingent valuation of recreation opportunities and environmental preservation stuff.

When's your paper due? Or are you just looking for junk to play with? I won't be in the office again until tomorrow. One of the girls in my cohort has some stuff on baby boomers and recreation preferences that has some good stuff (although you may end up with some miniscule R-squareds, depending on what you look at) -- it would probably work better than my thesis data, as you'd have to transform the crap out of my stuff. If you ever want to fart around with nonparametric data, though, I'm happy to shoot that your way, too.

Lemme know what you want and I'll try to hook you up.

You'd never guess that I'm probably doing a qualitative dissertation, would you? :evilgrin:

[bicentennial_baby]

23. My dataset doesn't have to be purely economic...

Samples of past projects are things like: factors affecting NHL revenue, role of outfield size in home run production in baseball, smoking race and low birth weights, waterfront property prices and erosion controls on Cape Cod, location choice for immigrants in metro areas, Supreme Ct appointments and party affiliation...

It's not due until the 2nd week in May, I need to have data collected by April 24th.

The Enviro/Recreation stuff sounds great! Or anything else! If you want my email, just shoot me a pm. Whatever you can give me would be awesome, I'm open to most anything, as long as it fits my requirements re: # of obs. and the type of variables.

:hug: You're the best

:)

[leftyclimber]

24. Check your PMs, sil vous plait.

:hug: