bob loblaw

I'm super interested in Bayesian Stats & this program so I have some insights. For full context, I have no interest in academia; I want to be an applied statistician/quantitative researcher role in government/industry. Point 1: UCSC is VERY Bayesian. Like people above have said, if you know pretty strongly that you're not interested in Bayes, UCSC might not be the place for you. Alumni comment: "Duke, for example, is known for being Bayesian. But they have a large Bayesian wing. At SC, the whole department is Bayesian". Point 2: I don't see any disadvantages of going to a Bayesian program for my career goals personally Even as someone interested in Bayes, I had some hesitance of going to a super Bayesian program. Here's a question I asked to an alumni who works as an applied statistician at RAND. Q: is there a disadvantage being in a Bayesian-heavy program when you become a statistician (in a place like RAND or the federal government) because a majority of statisticians are frequentist? A: Good question. I would say it is not a disadvantage, as a lot of places want to hire people who know Bayesian stats or are increasingly interested in using Bayesian statistics. However, it is true that a lot of my work ends up being non-Bayesian, and there are courses I wish I could have taken that UCSC did not offer (perhaps they do now?). These include causal inference and survey sampling/experimental design. These topics are very important at RAND or similar places, and were not a focus of the coursework at UCSC. Re: casual inference... they hired Richard Li (UW) last year who's research interest and background includes this area so perhaps they are filling this gap. As others have mentioned above, it may not be a disadvantage for academia either but I don't really know anything about this.

Yea I got an email from the program and also an official response through their portal. Don't lose hope. Also for those applying to UCLA Biostats, I got my rejection like last week. Just FYI

Sup y'all. I got my first acceptance today into UC Santa Barbara's Stats program! According to the survey page, some others did too. Wooooooooooooooooooo ?

Thank you @Stat Assistant Professor, @bayessays, @statsguy. This is all very helpful perspectives. I'm definitely not looking for entire summers off. From my perspective having 4-6 weeks off in the summer is incredibly long! That kind of time off doesn't happen when working in industry. I just wanted to get a sense of how summers work when you're a PhD student. @bayessays: if you're on a year-round fellowship, are you allowed any time off?

Thank you professor! Do some students in your program take a bunch of time off during the summers or is that uncommon?

I think for masters programs you have a pretty good shot at the top programs especially given you took real analysis and quite a bit of math. I'd even think that you would have a shot at some PhD programs as well.

Thanks everyone. Yea although highly unlikely, I hope one of NSF and DOE CSGF fellowships come through. @Geococcyx: Yea there's quite a bit of regional variation in hosting costs within CA as well. Ex: Berkeley is absurd, Riverside is cheap, etc. Regardless of where I end up, I'm very, very glad I have some savings (because I'll need to deplete them). ?

Just wondering what the typical summer looks like for a PhD Student: In year 1 (still completing coursework) In year 2 (almost done with coursework/quals?) In years 3 onward? I know some schools have different cultures around industry internships but wanted to get a sense of this. Thanks!

This question is for current students. I'm trying to do some budgeting since my income is about to drop significantly (once I start my PhD). If you can provide an very rough yearly average through out the 5 years that would be very helpful! Thank you. Glassdoor gave a estimate around $27,000 (which seems to match the number provided by OSU of ~$2300/month). But this could be on the high end? Also how much is your stipend when you're a first year student?

Does anyone know what the typical decision timeline is for Stats departments in the UC's (California)? I know it probably depends on the department but wanted see if anyone had any insight.

Thanks for the responses. Update: I confirmed with my program and they told me that especially given all the things happening this year, they completely understand if letters are late will accept them.

As a fellow Berkeley grad, I totally empathize with your "low" math grades (they're not that low). They're hardcore, their curriculum makes no sense, and they're horrible at teaching... Also going to Berkeley destroys most (if not all) of your self-esteem. ANYWAY, I think your grades in general are fine enough for masters programs. I also think you should submit your Stanford grades and you're being too paranoid. I think you'll be just fine. ?

So I am waiting for the third letter writer to respond to me/submit his letter. I've been in touch with them since September, requested the letter in early November, but they have ghosted me during these final weeks. My 2 other letter writers have already submitted theirs (on time). I finally got another academic to submit one for me (from electrical engineering) but he can't submit one for a few weeks. Just because the school allows up to 5 letters, I'm getting a supervisor from work to submit the 3rd rec (just so I meet the 3 letter requirement within a week of the deadline). I know late letters are not uncommon but how will this reflect on my application? I know some programs really prefer academic recommenders over professional ones.... I've turned in everything else on time...

Hey @StatsG0d, regarding the Real Analysis Class, yea I didn't have a ton of options because of Covid. But on the transcript, the course is indistinguishable from other courses at UIUC, which is a pretty good school so I think this is probably the best I can manage at the moment. Both my recs are from professors who are mathematically inclined (research on probability & manifolds) and I've been doing independent study with one of the professors for a full year so hopefully that would be a plus. Biostats + Stats: I dunno if it's not very well enforced but I've read on the admissions page for UCLA Stats that you can only apply to one program at a time. Did you not have that restriction?

Here's my profile! Just for context, I'm a few years removed from undergrad and have beefed up my proof-writing background/rigorous math background since graduating. ______ Undergrad Institution: UC Berkeley Major: Statistics || GPA: 3.85 Type of student: Domestic Male, in-state CA student GRE: Q-166, V-164, W-4.5 Relevant Courses During Undergrad: Math: Linear Algebra - non-proofs (A) Single & Multivariate Calculus (A's) Stats: Mathematical Stats (A+) Time Series Modeling (B+) Statistical Learning (A) Statistical Computing (A+) Linear Modeling (B+) Misc: Intro CS (A), EE course in Power Systems Engineering (A), Physics (A's) Post Undergrad: Math: Linear Algebra proof-based (B+, took it 2 years after graduation super rusty) Elementary Analysis (A - from UIUC) Independent Coursework: Real Analysis, Group Theory, Introductory Measure Theory with Oxford PhD Work Experience: 2.5 years as a quantiative analyst + data science internship Research: * 9 months in atmospheric science doing matrix factorization research * Working as a data analyst at the CA Dept. of Healthcare Services * 1 independent ML research project with well-known data scientist (PhD from UCSF) * 1 publication in a public policy journal as an undergrad. Letters: * Former Berkeley Maths Lecturer (DPhil from Oxford) * Former Visiting Stats Professor at Berkeley * 1 professional reference Research Interest: Bayesian Methods, Spatial Statistics, Environmental Statistics Programs I'm Considering: Ideal: UCLA (Biostats or Stats), UCSC, UCD (Stats) Other: UC Irvine, UCD (Biostats) Safety: UC Riverside, USC Keck

Hello! I was wondering if someone could advise me on whether to apply to Biostat or Stats given my weak-ish math background as UCLA makes you choose. The aim here is to try to maximize my admissions chances (obviously). I know Biostat applicants tend to be a bit less mathematically inclined. Here's my profile! Just for context, I'm a few years removed from undergrad and have beefed up my proof-writing background/rigorous math background since graduating. ______ Undergrad Institution: UC Berkeley Major: Statistics GPA: 3.85 Type of student: Domestic Male, in-state CA student GRE: Q: 166 V: 164 W: 4.5 Relevant Courses During Undergrad: Math: Linear Algebra - non-proofs (A) Single & Multivariate Calculus (A's) Stats: Mathematical Stats (A+) Time Series Modeling (B+) Statistical Learning (A) Statistical Computing (A+) Linear Modeling (B+) Misc: Intro CS (A), EE course in Power Systems Engineering (A), Physics (A's) Post Undergrad: Math: Linear Algebra proof-based (B+, took it 2 years after graduation super rusty) Elementary Analysis (A - from UIUC's NetMath program) Work Experience: 2.5 years as a quantiative analyst + data science internship Research: * 9 months in atmospheric science doing matrix factorization research * Working as a data analyst at the CA Dept. of Healthcare Services * 1 independent ML research project with well-known data scientist (PhD from UCSF) * 1 publication in a public policy journal as an undergrad. Letters: * Former UC Davis Professor (DPhil from Oxford) * Former Stats Professor at Berkeley * 1 professional reference Research Interest: Bayesian Methods, Spatial Statistics, Environmental Statistics Programs I'm Considering: Ideal: UCLA (Stats or Biostats), UCSC, UCD (Stats) Other: UC Irvine, UCD Biostats Safety: USC Keck, UC Riverside

Hey I'm applying to Stats PhD programs in the fall. So consider my advice with a lot of skepticism (I'm still trying to figure out stuff). Profile For masters, I think your profile is completely fine: I would think you'd get in most places. I would maybe consider taking more proof-based math courses in the future if you really want to do quantitative research. My understanding that showing that you can handle the rigor of a real analysis or proof-based linear algebra courses would move you up a level. Programs If I were you, I would maybe focus less on data analystics/data scienc-y programs if your goal is to do social science research. My impression that those are people who want to go straight into data science/tech. I think the MS at Duke would be awesome. It's a great combination of rigor & good applied classes https://stat.duke.edu/ms/ms-focus-areas-tracks/socialscience-policy Also would consider: "Our goal is to prepare students for PhD study in quantitative social science, and for professional positions at research institutions and government or nongovernment agencies. " https://mapss.uchicago.edu/areas-study/quantitative-methods-social-analysis https://gsas.columbia.edu/degree-programs/ma-programs/quantitative-methods-social-sciences https://steinhardt.nyu.edu/degree/ms-applied-statistics-social-science-research

So I'm increasingly more interested in biostats programs (over stats programs) and I came across this link: https://ww2.amstat.org/misc/BiostatsPhD2003-MostRecent.pdf I noticed USC (Univ. of Southern California) is listed on there and I didn't even know they had a program. Does anyone know anything about this program (reputation, etc.)? Their research emphasis on environmental health/preventative medicine seems interesting.

Because of Covid, it'll be really difficult to take the paper-based test anyway. I just finally got to take the general test recently and I think they closed down my testing center today lol. Even Stanford waived the requirement for this year https://statistics.sites.stanford.edu/phd-req-procedures

Agreed with everything above. I think your recent classes, a not-horrible GRE score (general), and your work experience would matter a lot more to them than your undergrad grades you got 10 years ago. Definitely highlight those aspects in your personal statement (if they apply for MS programs) as well.

Great suggestion. Btw, reached out to the chair of admissions at UCLA Biostats. They said it's not a requirement so as long as I can demonstrate that I understand the material, that's sufficient for them. They also said I should consider addressing this in my personal statement.

@StatsG0d First of all, thanks so much for responding. Also, I completely agree UW is a reach. I've definitely thought about dropping another $2k on a real analysis course as a non-matriculated student like I did with linear. It ain't cheap (esp. as an RA lol). However, I feel like with the schools I actually wanna get into (Davis & UCLA), I don't know if that will matter that much. Even if I get into UW for instance, I'd still probably go to Davis or LA. It seems to me what they want at the end of the day is mathematical maturity and the ability to handle the rigor. From this end, I thought using one of the letters to have my Maths prof vouch for me is probably the best option I have (he's fairly well-known and has a PhD from Oxford). I could totally be wrong about this so let me know what you think!

Great point. Forgot about that.

Btw, if you don't already know, I think being an in-state student may be a factor for the UCs. I talked to a very senior prof at Davis. She said that because I'm an in-state applicant, it'll be cheaper for the program to fund my studies than an out-of-state or international students. Of course, she didn't say explicitly say how much it affects admissions but she did say it "helps".

First, thanks for the quick response. Good to know that the Davis, UCLA, UCI, Colorado State are roughly in the same "tier". I guess with Colorado State, my experience on environmental science/clean energy is a really good fit with them that I figured I have a really good shot at getting in. ?‍♂️