STAT/BIOSTAT PhD Profile

[Lelouch Lamperouge]

Hi, guys,

 

I applied to some top Statistics PhD programs last year but got no offers. I decide to re-apply to some PhD programs again this year. I know that some people here are actually faculty members of some programs. I wish some of you could give me some evaluations.

 

I am an international student in U.S. college. I graduated from my university with High Distinction. I completed my undergraduate study in 3 years because I have transferred a lot credits from advanced placement.

 

My School: Top 5 Public Unviersity with high ranking in Stats

 

Major and GPA: Graduated with Highest Honors in Statistics (4.00/4.00) , Financial Mathematics (3.96/4.00)

 

Cumulative GPA: 3.93/4.00

 

 

Courses taken and taking: CAL III (A+), Diff. Equations(A+), Linear Algebra (A), Advanced Calculus (A-),Math of Finance (A-), Elementary Programming Course for C++ (A+), Intro to Theo. Stats (A+), Data Mining (A),  Applied Stats [graduate] (A), Statistical Computing (A+), Intro to Econometrics [under Econ department] (A+), Probability [graduate] (A), Independent Research in Stats (A), Honors Thesis (A), Discrete Stochastic process [graduate] (A+), Numerical Analysis (A).

 

GRE: 160(V) + 169(Q) +3.5(W) [ my writing is not good ]

 

I have not taken GRE subject Math; I think it is a weak point. I am going to take it in the end of Sept, but honestly speaking, my lack of learning some pure math courses makes me difficult to get high score from GRE subject Math. And I do not have much time to prepare for it.

 

Recommendation Letters: Three all from Stats Professors (1 from full professor, 1 from an associate prof, 1 from an excellent assistant prof). I was told by one of the professors (the full professor) that he wrote me a very strong letter. I guess other two are also good.

 

Research: have been doing a research work for my Honors Thesis, under a full professor's supervision, but with no significant result. Good at using R, Matlab.

 

Programs that I will apply to: 

      This year I will not apply to only the top 10 or top 15 Statistics PhD programs. I have not yet decided the list of the programs. I will still try most top 15 programs but the programs around top 25 are all acceptable.

      I will also apply to some Biostatistics PhD programs this year. Currently my goals are: Harvard, UMich, Johns Hopkins, UWashington. Are there any other very good Biostatistics programs that I could have a shot? Compared to the Statistics PhD programs, I am less familiar with the Biostatistics programs, so your recommendations would be very useful!

 

How do you think  my chance of getting any PhD offers (for Statistics or BioStatistics)?

 

Thank you in advance!

[DMX]

which schools did you apply to last year?

 

I am flabbergasted that you didn't get into any PhD programs--your profile is excellent. 

[Lelouch Lamperouge]

which schools did you apply to last year? 

 

I am flabbergasted that you didn't get into any PhD programs--your profile is excellent. 

Hi, I applied to virtually all the top 10 programs and some of my love places: Stanford, UCB, Harvard, UChicago, UWashington, UMich, Duke, CMU, Columbia, Cornell, Yale and Upenn.  

 

Since these programs are all with high reputation and very popular, I know the competition is quite high every year. This time I will apply to more programs and also include BioStats; hope I can get some offers.

 

Actually my professors were also surprised to my result last year but that was truth at least for the last time; wish not happen again this time .

[DMX]

Ah ok. I had similar stats to you (also international) but had problems cracking the top 15. Best of luck!

[Shostakovich]

I had nowhere near your stats (although I'm domestic and had some 'this guy is kinda unique' kind of things on the application) and still got into 3 of the top 5 for Biostat which I think are usually considered to be Harvard, UW, Johns Hopkins, UNC, and Michigan. Only Johns Hopkins was a straight rejection and I ended up applying to the MS at Harvard which I declined eventually.

 

One thing that might be a weak point in your app especially for Stats programs is the lack of Real Analysis. I'm not sure if Advanced Calculus refers to that, but I think it's pretty common for even Biostat applicants to have multiple quarters of it as well as multiple quarters of Linear Algebra. Since international admissions are very competitive, that may have knocked you down from the very top programs. Even so I think you're in very good shape for Biostats programs. I think they have lower bars for admission because (1) they get less applications (also less 'close to perfect' applications) and (2) they have more funding to take chances with somewhat 'borderline' applicants. UW publishes admission data (http://www.grad.washington.edu/about/statistics/summaries/2012/rptGsisAnnual_Dept2012.pdf) and it's pretty clear that you're competing with less people for admission in Biostats (261) compared to Stats (394).

 

Good luck!

Edited September 5, 2013 by Shostakovich

[Lelouch Lamperouge]

I had nowhere near your stats (although I'm domestic and had some 'this guy is kinda unique' kind of things on the application) and still got into 3 of the top 5 for Biostat which I think are usually considered to be Harvard, UW, Johns Hopkins, UNC, and Michigan. Only Johns Hopkins was a straight rejection and I ended up applying to the MS at Harvard which I declined eventually.

 

One thing that might be a weak point in your app especially for Stats programs is the lack of Real Analysis. I'm not sure if Advanced Calculus refers to that, but I think it's pretty common for even Biostat applicants to have multiple quarters of it as well as multiple quarters of Linear Algebra. Since international admissions are very competitive, that may have knocked you down from the very top programs. Even so I think you're in very good shape for Biostats programs. I think they have lower bars for admission because (1) they get less applications (also less 'close to perfect' applications) and (2) they have more funding to take chances with somewhat 'borderline' applicants. UW publishes admission data (http://www.grad.washington.edu/about/statistics/summaries/2012/rptGsisAnnual_Dept2012.pdf) and it's pretty clear that you're competing with less people for admission in Biostats (261) compared to Stats (394).

 

Good luck!

 

Thank you so much!

 

For the Real Analysis part, the Advanced Calculus in my school seems to be the Real Analysis in other universities (different names). But I am aware that the lack of Measure Theory is  my weak point for application to the Top STATS programs.

 

Thank you anyway!

[wine in coffee cups]

Just guessing, but it sounds like your problem could have been a lack of advanced math coursework? I see advanced calc with an A- and that's it. Was the graduate probability class measure theoretic or otherwise rigorous and proof-heavy? If it was, you might mention that explicitly in your statement of purpose and ask your recommenders to do the same.

 

If not, maybe you should take one or two more advanced math classes like measure theory or functional analysis before you reapply and do your best to get an A? A good score on the math GRE seems like it could help in your situation, but realistically, I think that's a longshot since you haven't taken algebra, topology, number theory, etc. while almost everyone else taking the test has.

 

Being a non-resident didn't help you either, but I also notice from your posts that you "write with an accent". When you applied last time, did you have your essays looked over by a native English speaker before you submitted them? Most applicants are not native speakers so some degree of grammatical errors is tolerable, but just want you to make sure what you sent didn't come off as sloppy.

 

Most importantly, you should apply to some larger programs. Most of the departments you applied to last year are small or medium sized, and all are highly ranked or at Ivy League universities that draw many applicants just because of the school name. (Kind of looks you used a name-based strategy last time around?) The biostatistics programs you're thinking about are all very selective, too. You ought to think about sending applications to some larger statistics or biostatistics departments like NCSU, UNC, and Iowa State, which are still well regarded and offer broad research opportunities but are big enough to make offers to more than the top couple dozen applicants.

Edited September 5, 2013 by wine in coffee cups

[Lelouch Lamperouge]

Just guessing, but it sounds like your problem could have been a lack of advanced math coursework? I see advanced calc with an A- and that's it. Was the graduate probability class measure theoretic or otherwise rigorous and proof-heavy? If it was, you might mention that explicitly in your statement of purpose and ask your recommenders to do the same.

 

If not, maybe you should take one or two more advanced math classes like measure theory or functional analysis before you reapply and do your best to get an A? A good score on the math GRE seems like it could help in your situation, but realistically, I think that's a longshot since you haven't taken algebra, topology, number theory, etc. while almost everyone else taking the test has.

 

Being a non-resident didn't help you either, but I also notice from your posts that you "write with an accent". When you applied last time, did you have your essays looked over by a native English speaker before you submitted them? Most applicants are not native speakers so some degree of grammatical errors is tolerable, but just want you to make sure what you sent didn't come off as sloppy.

 

Most importantly, you should apply to some larger programs. Most of the departments you applied to last year are small or medium sized, and all are highly ranked or at Ivy League universities that draw many applicants just because of the school name. (Kind of looks you used a name-based strategy last time around?) The biostatistics programs you're thinking about are all very selective, too. You ought to think about sending applications to some larger statistics or biostatistics departments like NCSU, UNC, and Iowa State, which are still well regarded and offer broad research opportunities but are big enough to make offers to more than the top couple dozen applicants.

 

Thank you for your suggestions! I will polish my essay this time by asking some native speakers to help me!

 

I also feel pity for my A- in Advanced Calculus, since I got full marks for the first midterm and the final but screwed up the second midterm. And it was in my third semester in college. The lack of more advanced math is my weak point so I will try my best in the GRE sub Math test, though I know I should not expect too much on that.

[qqyyzz]

I think your application is pretty strong.  I'm sure it will be stronger if you take more math courses of course but I think everyone here overvalues having graduate level analysis, measure theory etc on your application.  I think part of the reason you didn't get accept to some those programs is your lack of GRE subject test, not necessarily due to the lack of graduate math courses.  

 

If you had read through their admissions sites, I know stanford and UW's programs clearly state they require the subject GRE.  Uchicago, Columbia and I think Harvard "recommend" it.  A lot of times something like the subject GRE is just a departmental requirement and could mean automatic rejection just not having it.  However, I think not taking something like measure theory or other grad level math courses isn't an automatic rejection.  When I was visiting schools a few years ago, I'd say about 80% of the students I talked to didn't have many graduate math courses and many never had measure theory prior to entering the program.  The places I visited were not low ranked either, I'm talking about accepted students at places like cornell, wisconsin and UNC.

 

In fact when I talked to the graduate director at UNC, he had said that I had more math background than the average accepted student and in terms of analysis I had only had 1 semester of real analysis (undergrad), and two semesters of advanced calculus (also undergrad).

 

I was sort of in a similar situation when I was applying; I didn't take courses like number theory or topology or logic and as a result didn't have the required knowledge for the subject GRE.  However, a lot of programs realize that the subject GRE is not a great measure for grad student success and a lot of programs don't care if you haven't taken it.

Edited September 6, 2013 by qqyyzz

[Lelouch Lamperouge]

I think your application is pretty strong.  I'm sure it will be stronger if you take more math courses of course but I think everyone here overvalues having graduate level analysis, measure theory etc on your application.  I think part of the reason you didn't get accept to some those programs is your lack of GRE subject test, not necessarily due to the lack of graduate math courses.  

 

If you had read through their admissions sites, I know stanford and UW's programs clearly state they require the subject GRE.  Uchicago, Columbia and I think Harvard "recommend" it.  A lot of times something like the subject GRE is just a departmental requirement and could mean automatic rejection just not having it.  However, I think not taking something like measure theory or other grad level math courses isn't an automatic rejection.  When I was visiting schools a few years ago, I'd say about 80% of the students I talked to didn't have many graduate math courses and many never had measure theory prior to entering the program.  The places I visited were not low ranked either, I'm talking about accepted students at places like cornell, wisconsin and UNC.

 

In fact when I talked to the graduate director at UNC, he had said that I had more math background than the average accepted student and in terms of analysis I had only had 1 semester of real analysis (undergrad), and two semesters of advanced calculus (also undergrad).

 

I was sort of in a similar situation when I was applying; I didn't take courses like number theory or topology or logic and as a result didn't have the required knowledge for the subject GRE.  However, a lot of programs realize that the subject GRE is not a great measure for grad student success and a lot of programs don't care if you haven't taken it.

Thank you qqyyzz!

 

What if I do not have a good score in Sub Math? Is my chance to be admitted to Top 25 STATS program still good? How about the Biostats programs? Do they pay much attention to the GRE sub math?

[cyberwulf]

Things are pretty darn tough for international students these days. I think wine in coffee cups is right about your relative lack of advanced math courses being one reason you failed to crack the top stat programs; a strong Math Subject GRE score could help you on that front. Biostat departments will care less about your math prep, but they tend to value communication/writing skills more highly and also set aside a good fraction of their spots for domestic students, so the bar for international applicants remains quite high. Speaking for biostat, I think that places like Michigan, UNC, Columbia, Minnesota, Emory, Berkeley and Penn are decent options for you; you might not get into all, but I would be surprised if you didn't receive admissions offers from at least a few.

 

Have you considered doing a Masters degree first and using that as a stepping stone to a high-ranked program? Your record is such that you are a lock for admission at virtually every Masters program. Sure, you probably won't have guaranteed funding, but perhaps your letter writers could play their contacts for a suitable RA position? If you were to rock a top MS program, then combined with your undergraduate record I think you would find your chances of admission at good places dramatically increase.

[Lelouch Lamperouge]

Things are pretty darn tough for international students these days. I think wine in coffee cups is right about your relative lack of advanced math courses being one reason you failed to crack the top stat programs; a strong Math Subject GRE score could help you on that front. Biostat departments will care less about your math prep, but they tend to value communication/writing skills more highly and also set aside a good fraction of their spots for domestic students, so the bar for international applicants remains quite high. Speaking for biostat, I think that places like Michigan, UNC, Columbia, Minnesota, Emory, Berkeley and Penn are decent options for you; you might not get into all, but I would be surprised if you didn't receive admissions offers from at least a few.

 

Have you considered doing a Masters degree first and using that as a stepping stone to a high-ranked program? Your record is such that you are a lock for admission at virtually every Masters program. Sure, you probably won't have guaranteed funding, but perhaps your letter writers could play their contacts for a suitable RA position? If you were to rock a top MS program, then combined with your undergraduate record I think you would find your chances of admission at good places dramatically increase.

 

Hi cyberwulf,

 

Thank you!

 

I think I will also apply to the Master programs in Stanford and Chicago. I am also considering to apply to the master program in Harvard, but I have heard that the master program at Harvard is much more job-oriented.

 

Because you are a Biostats faculty, could you give me a list of names for the top 10 biostats program in your personal mind?

[cyberwulf]

I think the U.S. News provides a pretty good ranking, though you have to spend a few minutes distinguishing stat from biostat. Roughly speaking, the top 10 places are:

 

Harvard

Hopkins

Washington

UNC

Michigan

Minnesota

Berkeley

Emory

Penn

Columbia

 

I think you could also make an argument for departments like UCLA and Brown to show up in the last 2-3 slots.

[qqyyzz]

I think the U.S. News provides a pretty good ranking, though you have to spend a few minutes distinguishing stat from biostat. Roughly speaking, the top 10 places are:

 

Harvard

Hopkins

Washington

UNC

Michigan

Minnesota

Berkeley

Emory

Penn

Columbia

 

I think you could also make an argument for departments like UCLA and Brown to show up in the last 2-3 slots.

 

Hi Cyberwulf, I always hear about UCLA's biostat program, do you know anything about their statistics program?

[Shostakovich]

Hi cyberwulf,

 

Thank you!

 

I think I will also apply to the Master programs in Stanford and Chicago. I am also considering to apply to the master program in Harvard, but I have heard that the master program at Harvard is much more job-oriented.

 

Because you are a Biostats faculty, could you give me a list of names for the top 10 biostats program in your personal mind?

 

I heard Stanford's MS program is terminal, i.e. difficult to continue on to the PhD after the MS. Could be wrong though.

Hi Cyberwulf, I always hear about UCLA's biostat program, do you know anything about their statistics program?

 

I think their Stat program is a couple notches above their Biostat program, despite being a somewhat smaller program. They accept less applicants and have better placement in academia, unlike their Biostat which seems more geared towards industry.

[qqyyzz]

I heard Stanford's MS program is terminal, i.e. difficult to continue on to the PhD after the MS. Could be wrong though.

 

I think their Stat program is a couple notches above their Biostat program, despite being a somewhat smaller program. They accept less applicants and have better placement in academia, unlike their Biostat which seems more geared towards industry.

 

Well that's good to hear.  The only other mention of UCLA's stat program on this board is from a thread a few years old and the person did not have positive comments about their program.

[Lelouch Lamperouge]

I think the U.S. News provides a pretty good ranking, though you have to spend a few minutes distinguishing stat from biostat. Roughly speaking, the top 10 places are:

 

Harvard

Hopkins

Washington

UNC

Michigan

Minnesota

Berkeley

Emory

Penn

Columbia

 

I think you could also make an argument for departments like UCLA and Brown to show up in the last 2-3 slots.

 

I heard Stanford's MS program is terminal, i.e. difficult to continue on to the PhD after the MS. Could be wrong though.

 

I think their Stat program is a couple notches above their Biostat program, despite being a somewhat smaller program. They accept less applicants and have better placement in academia, unlike their Biostat which seems more geared towards industry.

 

Hi, Cyber and Shostakovich,

 

about the master program, how about the Michigan Master Program? Their master program seems to have a length of 1.5 to 2 years. Is the job opportunity good for UMich's master program after graduation? (I ask this questions for one of my friends who is considering to do a master program more geared towards industry.)

[cyberwulf]

Hi, Cyber and Shostakovich,

 

about the master program, how about the Michigan Master Program? Their master program seems to have a length of 1.5 to 2 years. Is the job opportunity good for UMich's master program after graduation? (I ask this questions for one of my friends who is considering to do a master program more geared towards industry.)

 

A Masters from any of the top 10 or so departments will be well respected in industry. 

[faerare]

I have another side question for cyberwulf about biostats Ph.D. applications. How many schools should one apply to? Is there a recommended range, or does it depend more on profile strength and domestic vs. international status? Thanks in advance. I have nine schools on my list currently, and they are a mix of sizes and rankings. 

[cyberwulf]

I have another side question for cyberwulf about biostats Ph.D. applications. How many schools should one apply to? Is there a recommended range, or does it depend more on profile strength and domestic vs. international status? Thanks in advance. I have nine schools on my list currently, and they are a mix of sizes and rankings. 

 

Suggest you start a new thread for this.

[wine in coffee cups]

I think your application is pretty strong.  I'm sure it will be stronger if you take more math courses of course but I think everyone here overvalues having graduate level analysis, measure theory etc on your application.  I think part of the reason you didn't get accept to some those programs is your lack of GRE subject test, not necessarily due to the lack of graduate math courses.

When I was visiting schools a few years ago, I'd say about 80% of the students I talked to didn't have many graduate math courses and many never had measure theory prior to entering the program. The places I visited were not low ranked either, I'm talking about accepted students at places like cornell, wisconsin and UNC.

In fact when I talked to the graduate director at UNC, he had said that I had more math background than the average accepted student and in terms of analysis I had only had 1 semester of real analysis (undergrad), and two semesters of advanced calculus (also undergrad).

I don't think Lelouch Lamperouge necessarily needs graduate level math coursework to be competitive, but rather a little more upper level undergraduate math coursework. You indicate you had two semesters of advanced calc plus one semester of real analysis, which might not sound like a lot to you, but that's still more than what LL was working with (one semester of advanced calc, and an A- at that). I think more applicants are coming from math major backgrounds than applied stat major backgrounds, too, so even if they don't have graduate coursework or measure theory, it's still common to have good grades in other upper-level courses like topology, abstract algebra, and complex analysis to vouch for their math abilities.

 

If you had read through their admissions sites, I know stanford and UW's programs clearly state they require the subject GRE.

Neither of UW's programs require the subject GRE. I don't think many students from the past few years in the stat department took it even when it was supposedly "strongly recommended". Perhaps expectations differ for international vs. domestic applicants, though.

 

Have you considered doing a Masters degree first and using that as a stepping stone to a high-ranked program? Your record is such that you are a lock for admission at virtually every Masters program. Sure, you probably won't have guaranteed funding, but perhaps your letter writers could play their contacts for a suitable RA position? If you were to rock a top MS program, then combined with your undergraduate record I think you would find your chances of admission at good places dramatically increase.

cyberwulf, I'm wondering if you might elaborate on why you think a master's is going to dramatically increase his chances? He's already shown he has great grades in undergraduate and graduate statistics coursework at a top 5 public with a good stats department, so his ability to handle the classes doesn't seem to be the concern. He self-reportedly already has good letters of recommendation from statistics faculty, too, and even some research experience. To me, it seems like he could take a measure theory class (or other advanced analysis, or graduate level stat theory) as a non-matriculated student and ace it to patch the only obvious hole in his profile for a much lower investment of time and money. Just curious what a 1-2 year master's program that's partially funded at best would meaningfully add when he appears to be already strong in the areas it could enhance.

[qqyyzz]

I don't think Lelouch Lamperouge necessarily needs graduate level math coursework to be competitive, but rather a little more upper level undergraduate math coursework. You indicate you had two semesters of advanced calc plus one semester of real analysis, which might not sound like a lot to you, but that's still more than what LL was working with (one semester of advanced calc, and an A- at that). I think more applicants are coming from math major backgrounds than applied stat major backgrounds, too, so even if they don't have graduate coursework or measure theory, it's still common to have good grades in other upper-level courses like topology, abstract algebra, and complex analysis to vouch for their math abilities.

 

That's a fair point, I did have all three of those courses when I entered grad school, but at the time I submitted my application I only had real analysis and the first semester of advanced calculus was in progress. 

[cyberwulf]

cyberwulf, I'm wondering if you might elaborate on why you think a master's is going to dramatically increase his chances? He's already shown he has great grades in undergraduate and graduate statistics coursework at a top 5 public with a good stats department, so his ability to handle the classes doesn't seem to be the concern. He self-reportedly already has good letters of recommendation from statistics faculty, too, and even some research experience. To me, it seems like he could take a measure theory class (or other advanced analysis, or graduate level stat theory) as a non-matriculated student and ace it to patch the only obvious hole in his profile for a much lower investment of time and money. Just curious what a 1-2 year master's program that's partially funded at best would meaningfully add when he appears to be already strong in the areas it could enhance.

 

My read on LL's disappointing results last year is that programs just didn't find enough in the application to get "excited" about. Doing a Masters gives you multiple opportunities to get faculty excited about your abilities: excellent performance in more advanced courses, top grades on exams, and outstanding contributions to research projects. It's probably best to look for a program where some Masters students go on to do PhDs (vs. one where everyone is going directly into industry), as these departments are more likely to provide the advanced coursework and exposure to research which will pique the interest of PhD adcoms.

[Igotnothin]

Interesting topic here, and nice to see cyberwulf's biostats rankings. In general I would recommend applying broadly, to a mix of stats and biostats. If you do a Master's first you will accumulate a lot of debt (maybe $70,000). This may help you, but there tends to be grade inflation in Master's programs, so even if you get a perfect grade point average it may only modestly improve your profile. Good luck and I hope you get into a PhD program this time. It is very unfortunate that it is so much harder for international students to gain acceptance to PhD programs.