Timeline for What should be offered in undergraduate mathematics that's currently not (or isn't usually)?
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 25, 2013 at 3:02 | review | Late answers | |||
| Jun 26, 2013 at 10:56 | |||||
| Aug 26, 2011 at 23:47 | comment | added | Thierry Zell | I see. It is certainly the case in the US that students in non-math programs (certain business, social sciences or information technology) end up learning a lot more statistics (at least in advanced degrees) than is offered to math students. The funny thing is that many such programs don't advertise it too openly, giving courses neutral names like "research methods". But of course, the statistics make students very employable in other contexts, see e.g. this New York Times article: nytimes.com/2009/08/06/technology/06stats.html | |
| Aug 26, 2011 at 21:20 | comment | added | Alex | Where I studied there was just math degree, no special statistics degree. And I didn't realize I'd do much statistics later. One thing about students is that they misplace their priorities getting fascinated by weird, fashionable or super-general stuff, and don't pay enough attention to the basics. That's where some guidance by the elders would be most appreciated. BTW, I'm not sure why one should separate math from statistics. To learn one but not the other is very dangerous for the future career. And in pure math randomness gives a different kind of intuition, just like geometry. | |
| Aug 26, 2011 at 11:32 | comment | added | Thierry Zell | Alex: shouldn't this be in a statistics degree rather than a math one? | |
| Aug 26, 2011 at 4:28 | comment | added | Alex | For example, how to compute confidence intervals when estimating many parameters? What if regularization is used? How to pick/estimate a prior? Some basic intuition about stochastic processes; an introduction to statistical machine learning; graphical models; how to model temporal data; how to estimate if you don't have enough data for asymptotic results to be valid; robust statistics; modeling interaction between multiple features, etc. Above all, some intuition that would make me feel how to proceed, what may work and how it may fail. | |
| Aug 26, 2011 at 3:44 | comment | added | Gerhard Paseman | I agree that there could be much improvement in how statistics is used as well as presented. However, most undergraduate curricula that (in my opinion) deserve the title have an offering in statistics, however elementary it may be. Was there a particular theorem or application you feel is not offered? Gerhard "Ask Me About System Design" Paseman, 2011.08.25 | |
| Aug 26, 2011 at 2:53 | history | answered | Alex | CC BY-SA 3.0 |