dogbones Posted September 3, 2018 Share Posted September 3, 2018 Hello all, I've been wondering where all of these RA jobs can be found? I checked NBER earlier today and only found 3 faculty looking for RAs posted... the odds seem very slim for someone looking to definitely find an RA job... are there other places to look? I imagine the Fed branches all take RAs, but do they have an official process to find and apply for RA jobs? Also, what are some of the think tanks, other than RAND and Brookings that actively recruit RAs? Thank you all... Quote Link to comment Share on other sites More sharing options...
zshfryoh1 Posted September 3, 2018 Share Posted September 3, 2018 There is a section on the NBER website for non-NBER RA jobs. This link should work Research Assistant Positions not at the NBER Some of the postings there are old and have probably already been filled. This time of year is a less busy time for postings since Spring is when most researchers looking for full-time RA's for the following academic year post, but new stuff gets posted as people receive funding or start new projects. A lot of top researchers post jobs there - at the moment the second and third posts are from Amy Finkelstein and Luigi Zingales respectively and in the past I have seen postings from Lars Peter Hansen, Raj Chetty, Emily Oster and Susan Athey. Quote Link to comment Share on other sites More sharing options...
dogbones Posted September 4, 2018 Author Share Posted September 4, 2018 zshfryoh1, that link is very helpful, there are so many postings! If not NBER, would there happen to be other opportunities? What is the "RA admit rate" approximately to the best of your estimation? And more specifically, how likely is it for a median qualifications level applicant to find an RA position? Quote Link to comment Share on other sites More sharing options...
tutonic Posted September 4, 2018 Share Posted September 4, 2018 From what I gather, the minimum GPA or cut-off happens at about 3.5/3.6. After you've crossed the hurdle, the thing that will land you an interview, and consequently the job will be past RA experience, proficiency in coding and interpersonal skills (which will be evident during the interview). The best way for a median quality candidate to get an RA position will be to go back to your alma mater and reach out to your old professors. That's usually how it goes for the majority of students. Pre-doc fellowship positions are extremely competitive (and rightly so). Quote Link to comment Share on other sites More sharing options...
mathenomics Posted September 4, 2018 Share Posted September 4, 2018 This time of year is a less busy time for postings since Spring is when most researchers looking for full-time RA's for the following academic year post. I don't think this is accurate- both fall and spring are peak-time recruiting seasons for RA positions. Most of the fall-recruitment positions should be posted throughout September and October, so you should probably wait it out a couple weeks. The best way for a median quality candidate to get an RA position will be to go back to your alma mater and reach out to your old professors. That's usually how it goes for the majority of students. Pre-doc fellowship positions are extremely competitive (and rightly so). I'd actually be surprised if there are that many paid full-time RA positions outside of NBER postings and Fed positions. Maybe the RA's on this forum can chime in. All the RA's that I know applied through pre-doc fellowships. Quote Link to comment Share on other sites More sharing options...
startz Posted September 4, 2018 Share Posted September 4, 2018 Another source of quite a few RA positions is JPAL and IPA. Some of them are very good RA positions for someone headed to grad school (but some aren't). Quote Link to comment Share on other sites More sharing options...
dogbones Posted September 4, 2018 Author Share Posted September 4, 2018 From what I gather, the minimum GPA or cut-off happens at about 3.5/3.6. After you've crossed the hurdle, the thing that will land you an interview, and consequently the job will be past RA experience, proficiency in coding and interpersonal skills (which will be evident during the interview). Would it possibly be well worth my while to study the intricacies of coding on places like stack overflow and read coding blogs, to become the best coder I can be, or is there some kind of a saturation point when it comes to coding for Economics? I'm thinking maybe it'd be good to be very good at coding since different kinds of questions can emerge that require coding skills? Quote Link to comment Share on other sites More sharing options...
dogbones Posted September 4, 2018 Author Share Posted September 4, 2018 I'd actually be surprised if there are that many paid full-time RA positions outside of NBER postings and Fed positions. Maybe the RA's on this forum can chime in. All the RA's that I know applied through pre-doc fellowships. That was kind of my line of thinking too. NBER and Fed are the main pools and then there's places like JPAL and IPA like startz mentioned... Quote Link to comment Share on other sites More sharing options...
applicant12 Posted September 4, 2018 Share Posted September 4, 2018 Would it possibly be well worth my while to study the intricacies of coding on places like stack overflow and read coding blogs, to become the best coder I can be No. Probably the only way to be good at coding is doing it. Reading others' code and think "oh of course that's how ABC is done in code. I can do it too" is like committing plagiarism, or like copying solutions to homework problems different kinds of questions can emerge that require coding skills? Unless you're a theorist, every question you encounter in economics requires coding skill Quote Link to comment Share on other sites More sharing options...
Rohanps Posted September 4, 2018 Share Posted September 4, 2018 No. Probably the only way to be good at coding is doing it. Reading others' code and think "oh of course that's how ABC is done in code. I can do it too" is like committing plagiarism, or like copying solutions to homework problems Disagree with this - if someone has written a package or made their code freely available online, of course it's fine to use that code (and just cite it in a footnote). Quote Link to comment Share on other sites More sharing options...
zshfryoh1 Posted September 4, 2018 Share Posted September 4, 2018 Disagree with this - if someone has written a package or made their code freely available online, of course it's fine to use that code (and just cite it in a footnote). I don't think applicant12 meant that using someone else's code is like plagiarism as far as being inappropriate/illegal. Rather they meant that by copying someone else's code or just reading a programming book without actually doing practical coding you don't really learn how to program similar to how a plagiarist never really understands the topic. Then when a new situation comes up, the wannabe programmer is unable to to code what they want and similarly the plagiarist is unable to do their homework/paper. Once you actually understand programming, then it is OK to copy someone else's code (with proper attribution) in order to save time and not re-invent the wheel Quote Link to comment Share on other sites More sharing options...
laborsabre Posted September 5, 2018 Share Posted September 5, 2018 If you are interested in getting better at R, you can check our the swirl() package which provides within-R exercises to increase your proficiency. You can then go onto Kaggle.com and do some of the great practice data projects on there. They use real-world data and there is tons of discussion if you get stuck. I would agree with applicant12 that the best way to learn coding is to do. Quote Link to comment Share on other sites More sharing options...
applicant12 Posted September 5, 2018 Share Posted September 5, 2018 I don't think applicant12 meant that using someone else's code is like plagiarism as far as being inappropriate/illegal. Rather they meant that by copying someone else's code or just reading a programming book without actually doing practical coding you don't really learn how to program similar to how a plagiarist never really understands the topic. Then when a new situation comes up, the wannabe programmer is unable to to code what they want and similarly the plagiarist is unable to do their homework/paper. Once you actually understand programming, then it is OK to copy someone else's code (with proper attribution) in order to save time and not re-invent the wheel Yes this is exactly what I was saying. Sorry for making my earlier comment too easy to misinterpret Quote Link to comment Share on other sites More sharing options...
tutonic Posted September 5, 2018 Share Posted September 5, 2018 If you are interested in getting better at R, you can check our the swirl() package which provides within-R exercises to increase your proficiency. You can then go onto Kaggle.com and do some of the great practice data projects on there. They use real-world data and there is tons of discussion if you get stuck. I would agree with applicant12 that the best way to learn coding is to do. I'd like to add by saying that if you're totally new to R, don't despair. It has a very steep learning curve at the start (bordering on overwhelming). Once you're past it, everything is smooth sailing. Quote Link to comment Share on other sites More sharing options...
dogbones Posted September 5, 2018 Author Share Posted September 5, 2018 If you are interested in getting better at R, you can check our the swirl() package which provides within-R exercises to increase your proficiency. You can then go onto Kaggle.com and do some of the great practice data projects on there. They use real-world data and there is tons of discussion if you get stuck. I would agree with applicant12 that the best way to learn coding is to do. This is very helpful, I did the first lesson in swirl and it's great because you learn R in R Studio... Quote Link to comment Share on other sites More sharing options...
dogbones Posted September 5, 2018 Author Share Posted September 5, 2018 I'd like to add by saying that if you're totally new to R, don't despair. It has a very steep learning curve at the start (bordering on overwhelming). Once you're past it, everything is smooth sailing. That's a sigh of relief for a future "me"... once I get there! Quote Link to comment Share on other sites More sharing options...
Kaysa Posted September 6, 2018 Share Posted September 6, 2018 R has a steep learning curve, but Datacamp flattens that curve. Datacamp is not free, but they give students a steep discount and you only need 2 months to learn the ropes. That said, I would recommend learning Python over R. Quote Link to comment Share on other sites More sharing options...
dogbones Posted September 8, 2018 Author Share Posted September 8, 2018 Hi Kaysa, I actually paid for one month of Datacamp after the free trial, and thought that it was too expensive and the videos weren't the greatest nor did it give as comprehensive an education as I could have with more time... On the other hand, thank you laborsabre again for swirl! It's allowing me to understand some parts of stackoverflow when I visit them for clarification, and I've even started a notebook for learning coding... can't wait to check out Kaggle.com too! And Kaysa again, my school uses R in graduate level Econometrics so I definitely have to learn it, but Python is my next language to learn after R. Do you use MATLAB at all? People on this forum have said that MATLAB is the best language for economists, but what are your thoughts? Quote Link to comment Share on other sites More sharing options...
Kaysa Posted September 8, 2018 Share Posted September 8, 2018 Hi Kaysa, I actually paid for one month of Datacamp after the free trial, and thought that it was too expensive and the videos weren't the greatest nor did it give as comprehensive an education as I could have with more And Kaysa again, my school uses R in graduate level Econometrics so I definitely have to learn it, but Python is my next language to learn after R. Do you use MATLAB at all? People on this forum have said that MATLAB is the best language for economists, but what are your thoughts? Hmm did you tell data camp that you are a student? I know some students that get data camp for free because of their student status. If you are using R for econometrics, I would read ISLR. The book is free, does an excellent job covering the linear regression and logistic regression, and most importantly, it contains code that shows the reader how to use R to run these regressions. It really makes things easy. If your econometrics course is going to use R for data cleaning, I would recommend learning Hadley’s tidyverse plus piping. It makes data cleaning much easier to do than base R. Matlab tends to be used mostly by economists that need to code their own empirics. I would not learn it unless you absolutely have to because you can usually do the same thing in R or Python, or even better, someone might have already coded it and deposited in github or cran Quote Link to comment Share on other sites More sharing options...
applicant12 Posted September 10, 2018 Share Posted September 10, 2018 I would not learn it unless you absolutely have to because you can usually do the same thing in R or Python, or even better, someone might have already coded it and deposited in github or cran Really? Quote Link to comment Share on other sites More sharing options...
laborsabre Posted September 10, 2018 Share Posted September 10, 2018 (edited) Hmm did you tell data camp that you are a student? I know some students that get data camp for free because of their student status. If you are using R for econometrics, I would read ISLR. The book is free, does an excellent job covering the linear regression and logistic regression, and most importantly, it contains code that shows the reader how to use R to run these regressions. It really makes things easy. If your econometrics course is going to use R for data cleaning, I would recommend learning Hadley’s tidyverse plus piping. It makes data cleaning much easier to do than base R. Matlab tends to be used mostly by economists that need to code their own empirics. I would not learn it unless you absolutely have to because you can usually do the same thing in R or Python, or even better, someone might have already coded it and deposited in github or cran My impression is that MatLab is used more for things like optimization/simulation. It is true that R and Python are generally capable of everything MatLab is, with the caveat that you just need to be more careful with R and Python because they are open source. All packages aren't created equal, and their performance isn't garunteed by a corporation like STATA or Matlab. Things aren't without tradeoffs, even if they are free. I'd still recommend R or Python over STATA or Matlab though. I know my school uses R, STATA, and Matlab. I'd also recommend not using base R for data cleaning - tidyverse is a good choice. So is data.table which is what I use (it is more intuitive for someone coming from STATA). Edited September 10, 2018 by laborsabre Quote Link to comment Share on other sites More sharing options...
Kaysa Posted September 10, 2018 Share Posted September 10, 2018 Really? Yes. MATLAB definitely does some things much better, but there is so much overlap between these three languages that it makes me question whether your time might be better spent on another activity rather than mastering MATLAB. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.