Most Productive Allocation of time

[IhorHrub]

Given someone has down time before they are starting graduate school, and they desire to study on there own so that they'll be most successful once school starts. What should the individual do? Do a bunch of math, read papers/books, learn R/SAS/STATA etc, or what? If someone has a model that shows how to optimally allocate time or even some nice robust econometric data of what things lead one to be most successful that'd be great.

[Food4Thought]

Are you trolling?

:

If no, then I would refer to my experience and the general consensus of many: don't try to study for your first year. Just try to relax and mentally prepare for a very demanding schedule.

 

If yes, then I would have to say that you have found a pretty gaping hole in the literature. I suggest you learn Stata and begin studying it. The data collection will be a little difficult, since you will need to send out many many many surveys to a bunch of incoming PhD's. Actually, try emailing your professors and ask what they did. That might give you historical data. Once you have that all sorted, you could run some sort of logit model and see how the various approaches influence tenure track placement at PhD granting institutions. And then you can learn Latex and type your results up and send it to a low level field journal. Don't worry, it will take some time for people to realize the masterpiece you truly have. Just remember, Akerlof's market for lemons paper was rejected from tons of journals before it earned him the Nobel. This shouldn't take you more than the summer, since you presumably got an A in real analysis. It will also be a great chance to buff up on your empirical skills, get that pesky dissertation out of the way, and be John Bates Clark worthy before you hit your second year.

[IhorHrub]

I was joking about the model or econometric research . I thought that was kinda obvious because it does sound ridiculous. I just want general advice.

[damomo]

Yeah, relaxing while you can is certainly good.

 

But, if you're dead set on doing something...learning Stata or R would be useful if you don't have any programming background/experience. Don't bother learning both, just pick one. It doesn't matter all that much. If you already have done some programming this probably isn't worth it.

 

I think going back to basics and going through some of the Khan academy videos for linear algebra, calculus, differential equations, ect. is a nice way to stay fresh and learn a bit without straining to hard. I enjoy it.

 

I think in another thread someone suggested reading through (at least intro and conclusion, preferably more) the papers of the people you think you might be working with. Likewise, skimming through some of the seminal papers in your field could be useful - both will make it a bit easier to get talking with your professors.

 

If you're the type who can just read through text books and actually get something out of the experience than working through Simon and Blume might be helpful.

 

But don't burn yourself out before you start - that's a recipe for disaster.

[IhorHrub]

Thanks I've going through textbooks and have been watching a decent number of videos, but I haven't really been doing to many problems as I'm just trying to grasp concepts. I just don't know if that's actually useful or not. I think I know R ok I've completed just about everything datacamp.com has to offer but I was still wondering if I should do more. Then for papers I feel that is enjoyable often but I feel not as productive as learning math or programming.

 

But yeah I likely am over stressing.

[mcsokrates]

If anything, reading papers is probably more important than cramming for math. At this point, you probably either have the math chops to pass the first year classes and quals or you don't. But there's 4 years of program after that, where you will largely not be proving theorems out of MWG.

[iamelben]

I've also heard some good advice recently: that now might be a good time to start tinkering around with LaTeX since it's used so heavily in paper-writing.

[Food4Thought]

I've also heard some good advice recently: that now might be a good time to start tinkering around with LaTeX since it's used so heavily in paper-writing.

 

Unless you're doing applied stuff. There are so many prominent empiricists that type things in Word. It's an abomination.

Edited May 9, 2015 by Food4Thought

[IhorHrub]

I've also heard some good advice recently: that now might be a good time to start tinkering around with LaTeX since it's used so heavily in paper-writing.

The R Markdown package I think basically does the same thing in making nice flashy documents except it's all in R,

[mcsokrates]

The R Markdown package I think basically does the same thing in making nice flashy documents except it's all in R,

 

So, 1) I think RMarkdown is cool! I use it a lot for early stages of work, and its super useful for quickly generating reports. I've heard that stats and some econometrics instructors have started nudging their students towards using it for homework. But, its not actually ready for primetime in terms of writing a full manuscript. LaTeX is still the best option here, for a number of reasons: 1) RMarkdown is great for integrating code that runs quickly, but its clunkier if you need to integrate code that takes hours or days to run (especially if that code needs to be run on a computing cluster). 2) LaTeX has packages to accomplish basically any typesetting task, while RMarkdown is still in relative infancy. 3) Journals (and your school's graduate school) will generally have required LaTeX style files.

 

That said, you should still be focusing on beers, beaches and relaxation. LaTeX, Rmarkdown, and programming in general is best done by actually doing. The gains to learning it now are relatively low.

[Kaysa]

Unless you're doing applied stuff. There are so many prominent empiricists that type things in Word. It's an abomination.

 

Why wouldn't they type things in word?

[mcsokrates]

I think substantially no one under the age of 35 uses Word to typeset papers anymore, right? You can tell whether a working paper was done in Word or LaTeX really easily, and the only new working papers I see that are obviously formatted in Word always have an older co-author.

[Kaysa]

I think substantially no one under the age of 35 uses Word to typeset papers anymore, right? You can tell whether a working paper was done in Word or LaTeX really easily, and the only new working papers I see that are obviously formatted in Word always have an older co-author.

 

I doubt it. In my graduating cohort, the only people that used latex were the theorists. It makes sense. Latex is superior to word when it comes to writing anything math. However, only a couple applied people used latex, while most used word.

[sulebrahim]

There are a lot of people who use lyX which is like a hybrid of word and latex.

I personally do very applied work but even I use latex. I had to keep up with the times.

[Kaysa]

There are a lot of people who use lyX which is like a hybrid of word and latex.

I personally do very applied work but even I use latex. I had to keep up with the times.

 

But why? What are the benefits you accrue versus the costs?

[MR14]

If you submit papers they need to be in certain formats; very easy to manage in Latex. That's the main benefit, especially for the bibliography.

[pch]

If you submit papers they need to be in certain formats; very easy to manage in Latex. That's the main benefit, especially for the bibliography.

 

But you can easilly do it with a citation manager like Endnote or Mendeley which integrate with Word, or is it something else you mean?