PhD in Statistics Fall 2021

[stats_sk]

Undergrad Institution: Top 50 in the US (known for math and stats)

Major: applied math and stats
GPA: 3.97

Type of Student: Domestic male

GRE General Test: Haven't taken yet, but I'm expecting 168+ in Q and 155-160 in V.
 

Programs Applying: Statistics PhD

 

Research Experience: Nothing impressive... but I've worked in a research-type role for several years after undergrad.
 

Letters of Recommendation: One of my college professors and two advisors/mentors from work. My advisor has a PhD in a quantitative field and will be able to talk about my research potentials.

 

Programming Skills: Python and SAS

 

Relevant Courses: all undergrad-level --- calc I-III, linear algebra, differential equations, abstract algebra, mathematical stats, probability theory (calc-based), stochastic processes, applied regression, and several more applied stats courses. I'm concerned that I have not taken analysis, but calc III, linear algebra, and differential equations were specifically for advanced math students and required us to learn how to do proofs (in addition to abstract algebra and stochastic processes, which were also proof-heavy).

 

Schools: Duke, Washington, Columbia, UNC, Michigan, Wisconsin, Penn State, Purdue, UCLA

 

My main concern is: Is my math background enough to get into these programs? Another possible path I can take is do a masters in math (or stats?) instead to build a stronger math foundation, but I wasn't sure if that's worth it if my ultimate goal is PhD. Thanks!

Edited September 1, 2020 by stats_sk

[Stat Assistant Professor]

I don't think you need to do a Masters in math or stats. However, I would recommend taking a semester of real analysis, since many schools do strongly recommend this. Ideally, you would be able to take real analysis this fall (or is it already past the add/drop period at your school?), and you can point out in your SoP that you are currently enrolled in Real Analysis when you apply. If not, maybe it will be just sufficient to state in your SOP that you plan to take Real Analysis in the spring, *but* you have taken other proof-heavy courses like abstract algebra and stochastic processes, so you have received training in doing advanced math and writing mathematical proofs. Also ask a LOR writer to point out that you have taken these advanced proof-heavy classes.

Of the schools on your list, I think Penn State, Purdue, and UCLA are "safely" attainable. Wisconsin also seems like a relatively safe bet. I could see you getting into Duke, Michigan, or Washington, but the admissions process might be a bit noisier at schools of that tier, so it's good that you're applying to several of these. Columbia is a reach but it doesn't hurt to try that one.

[stats_sk]

Thanks so much @Stat Assistant Professor. I should have made this more explicit -- I already graduated from college and have been working in an applied stats/data science-y role for several years. Good idea to ask my professor to specifically mention those proof-heavy classes. What I could do with real analysis is take the course through UIUC netmath (or something comparable) and mention this in the SOP - do you think this would help?

I'm relieved to hear that I have a good shot at schools like Penn State. You mentioned that Columbia is a reach, which isn't surprising. Would you say that schools like Cornell and Yale fall into the same tier as Columbia in terms of competitiveness? I guessed that even if they're outside of top 15, they probably receive a lot of applicants. Thanks again.

[Casorati]

46 minutes ago, stats_sk said:

Thanks so much @Stat Assistant Professor. I should have made this more explicit -- I already graduated from college and have been working in an applied stats/data science-y role for several years. Good idea to ask my professor to specifically mention those proof-heavy classes. What I could do with real analysis is take the course through UIUC netmath (or something comparable) and mention this in the SOP - do you think this would help?

I'm relieved to hear that I have a good shot at schools like Penn State. You mentioned that Columbia is a reach, which isn't surprising. Would you say that schools like Cornell and Yale fall into the same tier as Columbia in terms of competitiveness? I guessed that even if they're outside of top 15, they probably receive a lot of applicants. Thanks again.

You could take real analysis courses as a non-degree student in your undergraduate institution. Doing well in real analysis will improve your chances at top schools. Columbia/Cornell/Yale are competitive because of their general reputation and small cohort. However, their academic placements are not as good as top public schools like Washington/Berkeley/Michigan.

[Stat Assistant Professor]

I think admissions is very competitive to get into Cornell and Yale too. Of the 3 Ivies mentioned, I think Columbia and Yale are reaches, and Cornell is a slight reach.  I think Yale especially is a bit underrated relative to its most recent ranking in USNWR. This department has some heavyweights like Harrison Zhou, Andrew Barron, and John Lafferty, and it has placed some of its former PhD students in TT faculty jobs at UPenn Wharton, UC Berkeley, University of Chicago, as well as other good schools like UIUC, etc. 

[stats_sk]

Hi! I was reading through some of the older posts and saw that UCLA's stats program isn't as great as its name and ranking suggest. Does anyone know if UC Davis has a great program and if it's just as competitive as UCLA? Based on my profile above, should I consider it as a target/safety?

Edited September 5, 2020 by stats_sk

[Casorati]

UC Davis has a good program and is especially known for research in functional data analysis. I think you have a good chance of getting into there.

[stats_sk]

Hi All- I'm a non-traditional applicant. While working on my statement of purpose, I started wondering if my industry experience would help or hurt my chance of admission (or neither). If it doesn't matter too much, I would probably not talk about it very much in the SoP. Some specific questions I had were: 

1) Since I've been out of school for many years (5+), could the admissions committees be concerned that I forgot how to do advanced/pure math, despite my good grades from undergrad?

2) Would the committees think positively about the skills I gained through industry experience -- soft skills, "big picture" thinking skills, and the ability to manage deadlines? If so, would it be better if one of my LoRs discuss this, as opposed to the statement of purpose? If it doesn't matter for admissions, I wouldn't bother.

I would really appreciate to hear what you think. Thanks.

Edited September 6, 2020 by stats_sk

[stats_sk]

Hi - I'm considering applying for Biostats programs as well, but I'd prefer to be in a program that also focuses on theory, not just application. I have heard that UW and UNC cover theory, but are there any others? Assuming I take analysis in the spring, would schools like UW, Hopkins, UNC, and Michigan be a reach for me?

Thanks!

Edited September 19, 2020 by stats_sk

[bayessays]

Just now, stats_sk said:

Hi - I'm considering applying for Biostats programs as well, but I'd prefer to be in a program that also focuses on theory, not just application. I have heard that UW and UNC also cover theory, but are there any others? Assuming I take analysis in the spring, would UW still be a reach for me?

Thanks!

Given that from your grades you are clearly capable of handling analysis, I think just having it in progress will be enough for you - I don't think it's that important to have the grade. I think the first few schools on your list will be reaches (not crazy reaches, but far from guaranteed).  I think places like Penn State and UCLA should form a good chunk of the schools you apply to with a couple reaches and a safety. 

[stats_sk]

Just now, bayessays said:

Given that from your grades you are clearly capable of handling analysis, I think just having it in progress will be enough for you - I don't think it's that important to have the grade. I think the first few schools on your list will be reaches (not crazy reaches, but far from guaranteed).  I think places like Penn State and UCLA should form a good chunk of the schools you apply to with a couple reaches and a safety. 

Thank you for the quick reply. It looks like I edited my last post while you were responding. Are your referring to the schools I listed for Stats (in my original post) or Biostats (in my last post)? I'm definitely applying to Penn State for Stats.

[bayessays]

I meant in the original post. I think you can apply to any BioStatistics program reasonable and have a good shot at schools Michigan and below.

[StatsG0d]

53 minutes ago, stats_sk said:

Hi - I'm considering applying for Biostats programs as well, but I'd prefer to be in a program that also focuses on theory, not just application. I have heard that UW and UNC cover theory, but are there any others? Assuming I take analysis in the spring, would schools like UW, Hopkins, UNC, and Michigan be a reach for me?

Thanks!

No department *just* focuses on application. You will always have a Casella-Berger level of mathstats at any respectable department. The question is really how deep in theory you want to go.

UW and UNC have essentially identical curricula and are often considered to be the most rigorous. Similar to UNC, Harvard requires one semester in a foundational measure theoretic probability course. JHU also has a very rigorous curriculum, with a year-long sequence of measure theoretic probability theory. After the top-4, it falls off pretty quickly in terms of very rigorous theoretical training.

FYI, Michigan probably isn't the best place to go if you're interested in theory as they are a very genetics focused department. AFAIK they do not offer any measure theory at all, but I think you could take it in the stats department if you're interested.

[stats_sk]

 

7 minutes ago, StatsG0d said:

No department *just* focuses on application. You will always have a Casella-Berger level of mathstats at any respectable department. The question is really how deep in theory you want to go.

UW and UNC have essentially identical curricula and are often considered to be the most rigorous. Similar to UNC, Harvard requires one semester in a foundational measure theoretic probability course. JHU also has a very rigorous curriculum, with a year-long sequence of measure theoretic probability theory. After the top-4, it falls off pretty quickly in terms of very rigorous theoretical training.

FYI, Michigan probably isn't the best place to go if you're interested in theory as they are a very genetics focused department. AFAIK they do not offer any measure theory at all, but I think you could take it in the stats department if you're interested.

Thank you! I didn't know much about Harvard and JHU. Thanks for the insight on Michigan.

 

[stats_sk]

1 hour ago, bayessays said:

I meant in the original post. I think you can apply to any BioStatistics program reasonable and have a good shot at schools Michigan and below.

 

Thanks. Just to clarify, are you saying that I have a reasonable shot at places ranked above Michigan? I was aware that Biostats programs were less competitive than Stats for domestic applicants, but I didn't know how much math background is usually needed to get into those top schools. I'm wondering if it'd be a waste to apply to top 3.

Edited September 19, 2020 by stats_sk

[bayessays]

I'm saying your profile is good enough where you will be in the conversation at top places and it would be worth applying - you probably have an average math background for the top places. But the top programs are very competitive so I would not count on them.