In which I recognize that Hardy (the Hardy Weinberg model, teacher of mathematics at Cambridge) inspired all my Phd thesis, because he was used to say to his students that
"mathematics is beautiful precisely because it is useless"
so, my Phd thesis is the most beautiful work I ever see.
Subscribe to:
Post Comments (Atom)
catastrofica come al solito!
ReplyDeleteRiccardo dice catastrofista di solito. non catastrofica.
ReplyDeleteKATA: giù in greco
STHROPE: volgo in greco
= capovolgo
ah tony grazie di tutte le tue mail piene di consigli utili.
...beh... lo spirito e' quello giusto!
ReplyDeletevedere il bicchiere mezzo pieno! :D
Oh, at least one optimist!!! :-)
ReplyDeletee poi io ho deciso che la mia tesi sarà necessaria per scoprire il bosone di Higgs... siamo un esercito di sognatori, per questo siamo invincibili! (cit.)
ReplyDeletemoreover I've decided that my thesis will be necessary to discover Higgs boson... we are an army of dreamers, thus we are invincible! (cit.)
Cos'è il bosone di Higgs?
ReplyDeletehttp://lmgtfy.com/?q=bosone+di+higgs&l=1
ReplyDeleteSorry guys, but I don't have the feeling that my work is useless... I'm trying to do, instead, something useful, something that could be used in real applications. Maybe, eventually, it won't be used, but this is anoher story... it's sad that you think you're doing something useless... I told you: you all should have moved to the amazing world of permutations, or at least who feels useless!!! Marlies!!! I told you!!!
ReplyDeleteNo polemica, semplice osservazione!
Ciao bei e bea
Checca
You, not me, think that your work won't be used. I'm confident that physicians will use integrated likelihood: http://cdsweb.cern.ch/record/1099980/files/p125.pdf
ReplyDeleteI don't see any permutation model proposed for the signal noise detection, and it is an observation, not a polemic. :-)
My work is useless: no one in the real world uses likelihood methods for panel data. All people use method of moments, which is really a good choice because it is simpler and you do not have to invert any matrix. I have done a choice, I do not want to work anymore with prof salvan and co. after the Phd, even if there would be the possibility. I just want toi maximize the distance between me and any form of ipocrisy...
ReplyDeleteAlways the same story.
ReplyDeleteResearch is not always useless. Maybe your ML panel data is useless for panel data analysts now, but in 20 years maybe someone at NASA will use your method for better a undertanding of signals from space crafts or whatever...
The same with permutations. Nobody in real life use Checca's method even if she think her method for real applied people. But somethimes someone will use them... (maybe... I don't think so...this is clearly polemica!!!)
Research seems useless at a glance but is not so. Ok something is really useless, but if you do research your not maximizing 1) your salary (as previous post underlined) 2) pragmaticity of what you do.
You wanna do Statistic with immediate utility? Go to a research center and analyze their data with simpler tools and give them quick answers (yes even the t-test!), go to Wall Street and let your company earn thousands of dollars with coherent forecast,or go to google and.... control the world!
I wrote a poster before and now I removed it because I was so angry, sometimes I overanalyze facts...anyhow Tony you are right, but finishing the Phd, I want to do something with immediate utility, as suggested by Nitzsche. Ciao
ReplyDeleteI'm quite, but not totally agree with Tony. From a theoretical point of view, what he said is perfect, but practically you know that there are topics, e.g. latin, that are important for the culture, but without a real utility: you know how much I love latin, but I see it more as an hobby for a retired researcher than as a topic in which you have to train and fund a dedicated researcher. Probably, during a slump, it is preferable to cut this kind of research than to delete one scholarship for any phd schools... and to leave this kind of topic for a more positive economic time...
ReplyDeleteBut what I wanted to said? Something like 'it is ok to choose a topic without an immediate utility (as Tony said, we are researchers, not analysts), but think if it could be useful (e.g., I hope that nobody of you will research the formula to do up the shoes in the faster way...)'
An why not something important for the culture which is also useful??!This is the ideal
ReplyDeleteOk guys...
ReplyDeletethe interesting topic are only when I don't check the blog... I begin to think there's a causal effect...
Anyway, I think theoretical research is useful... Even if it would never be applied... I think culture is important by itself... Anyway, I kind of agree with Riccardo that if you want to cut funding you should start from research that doesn't give you money back... it's sad but I agree..
Anyway, you know what I think about statistics... I like to start from a problem and try to solve it. And if you can solve it with a 1920's Fisher method why not? But if you can't, then you can develop a new method! There are a lot of *real* problem waiting for a solution!
This is the way they think here in Berkeley, anyway... and I love it!
On the other hand, mathematics is so interesting, so I can't say nothing bad about who likes theoretical statistics...
Maestro, you're always the maestro! Toni, what you say is true, you can work today on something which is useless now, but which could solve perfectly an important problem in 20 years. So Marlies, don't worry. But what Marlies is underlining is, I think, that she doesn't feel the importance and utility now of what she's doing, and this is something she's missing a lot. For me it's the same, I could never work on something without seeing its applicative utility, and for which then I have to spend weeks looking for an real data set which could be successfully analyzed with my method...(and here we go back to Maestro's idea: "I like to start from a problem and try to solve it").
ReplyDelete