- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: SEC schools with good Statistics degree programs.
Posted on 12/27/16 at 10:38 pm to UF
Posted on 12/27/16 at 10:38 pm to UF
Yeah medicine is really tough to break into with new techniques or revelations. They are stuck in a paradigm that's been pretty reluctant to switch over.
In fact most hard science fields are in the same boat.
The future is going towards computer science or "big" data.
Moving towards Bayesian as well
In fact most hard science fields are in the same boat.
The future is going towards computer science or "big" data.
Moving towards Bayesian as well
Posted on 12/27/16 at 11:02 pm to gaetti15
quote:
The future is going towards computer science or "big" data.
Moving towards Bayesian as well
Bayesian yes.
Big data - I think no. We are reducing the complications in data by producing charts, graphs, and summary statistics. But insight is not being produced. As long as one's goal is to look at the marginals, big data is good. If you want to dig into the conditionals, it fails horribly for a myriad of reasons such as sloppy data entry, measurement error, etc. Those things don't affect the fact that a majority of people pick X, but they usually break down once you want to figure out P(x|y) and so forth.
There was an article, in Proceedings of the National Academy of Sciences (PANAS), I think, earlier this year, that gave a nice breakdown of this topic.
This all has its genesis in the aftermath of WWII. An attitude developed in the US that if we could just create enough computational horsepower, we could figure out anything. But it continues to fail us. It started becoming evident in the 60s, turned into a crises in the 70s, note Carter's 'crisis of confidence' speech and all that lay behind the genesis of that speech, and things just continue to move in that direction. In essence as a society we decided to double down on computational horsepower as the key to our future and we're just not seeing the returns as anticipated, or even close.
Most statisticians do understand this. But their customers just want the results, the graphs, tables etc. They don't like to hear the analysts explain all that nasty uncertainty.
This post was edited on 12/27/16 at 11:04 pm
Posted on 12/27/16 at 11:09 pm to stat19
if you put in the work, you can achieve greatness at any sec school, at least in undergrad. My brother has a statistics degree and masters from uga and ended up at Harvard for his Ph.D. in statistics. He does alright now to say the least.
Posted on 12/27/16 at 11:20 pm to rockchlkjayhku11
quote:
if you put in the work, you can achieve greatness at any sec school, at least in undergrad. My brother has a statistics degree and masters from uga and ended up at Harvard for his Ph.D. in statistics. He does alright now to say the least.
did he work directly under robins or hernan there at harvard? Those are the two names I've followed most closely there, along with rubin to a much lesser degree.
Posted on 12/28/16 at 12:06 am to UF
No fricking clue but after a quick google, my guess is no to the first two and a completely unsure "probably" to Rubin.
Posted on 12/28/16 at 12:13 am to stat19
Mostly about 83% of the schools in the SEC
Posted on 12/28/16 at 12:13 am to UF
quote:
UF
You seem to know your line of work...maybe Chicken will let you on the Tiger Rant--they need some help over there with getting some of LF7's stats correct...
Popular
Back to top

1




