statfan

Thanks for your input. The first two letters should be fairly strong since they are very familiar with my work but I am not sure whether I should get my third letter from a pure math lecturer or a well-known biostatistican. The pure math lecturer knows me very well but I'm wondering if a letter from a lecturer (he has a phd though) would end up not helping me. I feel a letter from the well-known biostatistican is likely to be generic.

Hey I will be applying to statistics/biostatistics phd this fall and I'd like to know where I stand among other applicants. Any evaluation/advice is greatly appreciated. Here is my profile: Undergrad Institution: One of the big Math/Statistics schools in Canada (one of U of T, U of Waterloo, UBC) Major(s): Statistics and Actuarial Science GPA: We do percentage grading. Roughly convert to 3.7, higher in math/stat courses Grad Institution: Same school Major: Statistics GPA: 3.9+ Type of Student: Asian Male Relevant Courses: Calculus 1-3 (A+, A+, A+). Linear Algebra 1-2 (A+, A+). Introduction to Differential Equations (A+). Introductory real analysis (A-). Real Analysis (A-). Lebesgue Integration and Fourier Analysis (A+). Mathematical Statistics (a la Bain/Engelhardt, A+). Estimation and Testing (A+). Graduate Statistical Inference (A+). Applied Probability (A+). Stochastic Processes (A+). Applied Linear Models (A-). Generalized Linear Models (A). Simulations (A). Time Series (A). Sampling and Experimental Design (A+). Experimental Design 2 (A+). Survival Analysis (A). Biostat Topic: Missing Data and Causal Inference (A+). GRE General Test: Q: 170 V: 153 W: 3.5 Should I retake due to the low verbal and writing scores? GRE Math Test: Taking in September Research Experience: One part time RA with a Statistics professor during undergrad. Graduate RA under the supervision of a biostat associate prof and now I'm working towards my master's thesis. I will finish my thesis by the end of this year. If things go well, this will result in several publishable papers next year. Awards/Honors/Recognitions: Something related to high grades, nothing major. Letters of Recommendation: One from my current master supervisor (associate prof), one from an associated prof who I took grad stat inference and experimental design with. I am not quite certain about the last one yet but it will be either from a pure math lecturer who taught me lebesgue integration that I did very well, or a well-known biostat prof who taught me survival analysis. Any Miscellaneous Points that Might Help: Although I did well in most math/stat courses, I received some low grades (B's and two C's) in actuarial science courses and electives. How will this negatively impact my chances? Should I address this in my SOP? Applying to Where: This is my tentative list and it is subject to change depending on my GRE math results. Statistics: Stanford, Berkeley, Chicago, Penn, Wisconsin Biostatistics: Washington, Johns Hopkins, Michigan, Columbia Thanks for your help!

Real analysis is very useful for statistics at the phd level and people seldom get into statistics phd without having taken real analysis. Probability and mathematical statistics are important as well but this background can be made up during phd. For MS calculus 3 and linear algebra are more than enough.

GRE is a filter and unless you bomb it, it won't matter much. Your official transcript should not show your GPAs since different universities have different rules of calculating GPAs. The admission committees will mainly focus on grades in individual courses that are relevant and you did well in all of them. That being said, you are competitive for any masters program so maybe remove some lower ranked schools like Ohio state and Colorado, and I feel that with this profile you are even competitive for top 20 phd progams.

I doubt they would care about your major gpa since your major is unrelated to statistics. You have more than enough background for masters and with this profile you should be competitive almost everywhere.

U of T biostat department is pretty much unknown and UBC has a very small department. Mcgill biostat is highly regarded in Canada and so is Waterloo.

You have a very strong profile coming from a well-known school, so you should apply to mostly top shcools. The UC programs you listed are far removed from top ones and are not competitive as you thought. Your original list seems good and maybe you should apply more aggressively by removing schools like Vanderbilt and JHU since they don't have a very strong department. I am sure you definitely can get into some top 10 programs, if not all.

Your odds depend on the school you attended and the letters. There is a big difference between Toronto/Waterloo and say Western. Sometimes the decision really comes down to one single letter.

It may be due to weakness in other parts of their applications. It is very common that strong applicants have several phd-level statistics courses and I personally think it helpful to get exposed to grad courses early to see if graduate school fits you. Another advantage of doing this is that you can move onto research faster than others.

Many students who went to top schools in the US in my school had taken multiple graduate level stat courses such as probability theory/statistical inference. I agree that real analysis/linear algebra are very important but other math courses are much less related to statistics. It may be a good idea to go to stat masters as a stepping-stone to phd and make up the real analysis background.

I disagree with this. Doing well in grad level statistical inference and/or probability serves you two purposes. (1) It shows your aptitude for grad level research. (2) Grad classes tend to be much smaller and it is easier to establish close relationship with your prof. This can result in a great letter from a well-known faculty member cuz grad course instructors tend to be more senior. Taking grad level analysis courses would definitely help too and if you do well in them, you will be well ahead of most applicants. However, this does not necessarily requires a math master. Taking abstract algebra/geometry certainly wouldn't hurt but under time constraint, you would want to focus on the most important things.

Your grades should be fine for the schools you applied but your math background is relatively light compared to strong applicants. You would want to have at least one mathematical statistics in your transcript, perferrably two. Maybe the problem lies in your letters. Generic letters talking about how well you did in their classes won't help for phd application, instead you would want to have letters talking about your research. Also, having one or two amazing letters from top faculties could dramatically improve your chance.

The admission process is very complicated and no single factor plays a decisive role. GPA is only meaningful given certain context. A lower gpa may be more impressive at a prestigious instituition than a perfect gpa from an unknown institution. People from lower-ranked instituition will have a much tougher battle. Also, your math/stat gpa matters much more than the overall gpa. What is your math background? How well did you do in each course? For example, if your B's are in real analysis/linear algebra/math stat, that may be unfavourable. Your recommendation letters play an even more important role than the GPA. If you get strong letters from eminent faculty members highlighting your research potentials, that would really help you. I agree with bayessays that your verbal score is so low that it would raise concerns about your ability to communicate in English. Though verbal section of the GRE is one of the less important aspects of the whole application but still you do not want to stand out in a bad way. I would say at least aim to get 50th percentile in order not to be screened out.

Your math/stat gpa matters much more than your cumulative gpa and unfortunantely your major gpa turns out to be much lower than your overall gpa. Although masters are much less competitive than phds, your consistent B's in math courses is still a huge concern. Without evidence of strong math skills in your other part of application, it would be unlikely to get into top masters programs. I would advise you to look into some programs ranked 20-30, or do well in math GRE to show that you actually have sufficient math skills.

Only Stanford requires the math GRE. UW and Columbia recommend it. However, neither the general GRE nor the math GRE plays a decisive role when it comes to admission. Math/stat background and recommendation letters play a much more important role than the GREs.Time is a fixed constraint and you should prioritize the things that matter most. I wouldn't sweat too much on the GREs but instead focusing on research and upper-year/grad level math/stat courses.

You are missing the point here. GRE is a filter and they only care about the quantitative section. You should score high in quant in order not to be weeded out during the screening process. You would be fine if you do not bomb the other two sections. Except a few schools, others do not care about the math GRE. That being said, do not submit the score unless you are very confident of doing well. Also, OP did not ask anything about the application fee, it's up to OP to consider this and I don't understand why you brought this up.

What is your math background and how did you do in your math/stat courses? I am asking because this is much more relevant than your overall gpa. If you did uniformly well (mostly A's) in your math courses, then you may stand a chance of getting into good schools. Otherwise, admission committees may doubt your ability to do actual math given your low gre score and you may be weeded out in the screening process in some schools.

Thank you very much, this is much more reassuring as I got mostly A+ and A's in my core math and stat courses. I have a few very low grades in finance/actuarial science though. I just done my real analyisis exam and I don't feel that I did very well. I am afraid that I may end up with high 70s or low 80s. In many Canadian universities, we do percentage grading, so that would correspond to B+ or A- (77-79 B+ 80-84 A- 85-89A 90-100 A+). Should I take Lebesgue Integration to remedy this? Would admission committees refer to the grading scheme on transcript to interprete the percentage grade as high 70s/low 80s does not sound impressive at all.

When you go over the transcript, how would you assign weights to courses from different disciplines? (eg. from 1-10 scale with 10 being the most important) 1. Calculus/Linear Algebra 2. Real analysis/Measure theory 3. Other pure math courses (eg. geometry/abstract algebra) 4. Undergraduate statistics courses 5. Graduate statistics courses 6. Courses from other quantitative disciplines (eg. actuarial science/physics/chemistry/finance) 7. Electives 8. CS courses

Few masters fund students. The masters that fund students are generally research-based and most students finally pursue a phd. I heard penn state and florida have a funded master.

You have solid grades in a number of proof-based math courses so occasional slip up won't matter much. Many applicants do not have any research experience, so you definitely have an advantage over them. Along with recommendation letters from well-known professors, I think you can get into most schools you listed, if not all. Actually, you are way under applying, probably include some top schools in your list. I'd say you stand a good chance at top 10 stat phd.

I don't think that you need TOEFL score if you did your undergrad in the States, even if you are not done. You have high grades in upper level math courses so your occasional lower grades won't matter much. Most departments don't care about your GRE verbal and writing unless you really bombed it. 154 verbal is a respectable score and you will be fine with that score. Your overall profile is very strong and I think your lists are fine. Most top phd programs admit a lot of international students, what they really care about is that if you have the potential to make contribution in this field. That being said, they would prefer a qualifed international student than a mediocre domestic student.

88% is a respectable score and I doubt that it will hurt your chances, actually it may help you if you come from a lesser-known school. It is not true that admission commitees expect higher gre math scores from international applicants. The fact that international students have higher scores than domestic students is simply because those who scored poorly did not submit the result.

I'd say do not worry about several lower grades too much as your overall math grades are good. If you look into masters that would be helpful to your further application to phd, I advise you to look into some programs like UBC, Toronto, Waterloo in Canada as they are all research-based masters. The class size tends to be much smaller than that of US masters since you will be fully funded. This makes it easier to develop a strong relationship with professors. You can also choose to take some advanced stat courses as master/phd courses are held together. With that said, these programs serve as a stepping-stone to phd, and will certainly boost your chances significantly if you do well in the program. Most course-based masters programs in the US won't help you much when you apply to phd.

Your math background and grades? This is important cuz it is much more relevant than your overall gpa. If you have courses through multivariable calculus, linear algebra and a few statistics courses with decent grades in general, you should be competitive in most masters programs, including Columbia. Masters in the States are unfunded so the bar tends to be much lower, with Columbia's extremely low as they admit tons of students into masters. However, 161 Q is a bit low even for masters, I advise you to boost it to at least 166.