Garth

Home/Garth

About Garth

This author has not yet filled in any details.
So far Garth has created 31 blog entries.

Talk to new PhD students

Today I was asked to give a talk to new research students in the School of Maths and Stats about my experiences. Here’s what I prepared:

I’m going to talk about three things:

1) daily struggle of being a PhD student
2) not overloading yourself with non-PhD related commitments
3) conferences

1) Throughout undergraduate life you tend to be successful pretty regularly. Your work is doled out in bite sized pieces, you get positive reinforcement from doing well in an assignment/quiz/exam. At the end of the semester you got a set of marks, probably Ds or HDs, to let you know that you are on track.

As a postgraduate student that doesn’t really happen. You’re not going to be successful on a daily basis. There’s a good chance you won’t even be successful on a monthly basis. There will be a time when you question your life choices – where you wonder why on earth you chose this path. For weeks on end you might find yourself banging your head against a wall.

And just when you’re about to throw in the towel and give up,
– a spark of an idea will come to you,
– or you’ll find a journal article that gives you a new perspective on how to solve the problem,
– or you’ll find new meaning in an old article that was utter gibberish when you first read it a few months ago
and you’ll get a little jolt of ecstasy as that Gordian knot of a problem you’ve been working on unravels before you.

Those are the good days – you’ll get a rush of endorphins as a present to you from your body for a job well done and you’ll take the afternoon off and go home with a smile on your face. You will remember those days – they are what gets you through the leaner times as you set your mind to the next problem and once again start the daily slog to the next watering hole.

I don’t think there’s any way to avoid the slog, your supervisors might be able to give you some directions to help you navigate the wastelands, but they don’t have a complete map either and there’s no guarantee that they’ve pointed you in a fruitful direction.

Research takes time. Which brings me to my second point.

2) It helps to have a lot of time to devote to research. This may mean limiting your extracurricular activities when they take too much time out of your regular research day. For me, I did way too much teaching early on in my degree – as I was teaching here and in the Business School. I wouldn’t go so far as to say that I regret doing all the teaching – I enjoyed it – but it definitely slowed down my progress. For other people it’s trying to do a PhD full-time while working a couple of days a week elsewhere – they almost invariably find that one or both suffers as a result.

If you like teaching, you can apply to become a postgraduate teaching fellow – applications usually open up at the end of the year – it gives you a fancy title, a regular pay cheque, the possibility of a bit of lecturing, and it doesn’t overload you with too much work.

Life is also more enjoyable when you don’t overload yourself. When I had time, one of the high points of my day was taking a paper and a highlighter down to the benches between Carslaw and Madsen, sitting out there in the sun reading, highlighting and contemplating. Just thinking about stuff.

The final thing I want to talk about is going to conferences.

3) I’ve had the opportunity to go to a number of domestic and international conferences, both specialised conferences with around 100 people and larger more general conferences. As soon as you have something to talk about, I’d suggest going to a specialised conference – this probably means you have to (get to) travel overseas.

I went to a specialised conference about 18 months into my degree. It was ICORS in Spain and I had an awesome time.

I didn’t really appreciate how important it was for my development at the time – but looking back now I think it was pretty important:
– EXPOSURE I was exposed to a heap of different ideas, while not directly relevant, they did give me a broader understanding of all the different areas in my field which made reading articles easier.
– NETWORKING I met a lot of people, so when I went back a couple of years later, I already knew people there or knew their colleagues, feel like you’re a part of a larger community. Also putting names to faces on journal articles makes them more interesting – especially if it’s tied to a fond memory of a tapas bar crawl or 3am beers in the bar of a Russian hotel.
– FEEDBACK And perhaps most importantly, it forces you to write-up your ideas and present them to a potentially critical audience so you can get some feedback (other than relying on your supervisors)

There are lots of funding opportunities if you know where to look,
– PRSS (postgraduate research support scheme)
– your research group usually has a bit of money for conference travel
– there are additional scholarships occasionally advertised on scnews (the School’s electronic notice board)
– professional bodies such as SSAI or AustMS sometimes have opportunities too – to take advantage of these you usually need to have been a member for at least a year. It’s $20 a year for SSAI and student membership of AustMS is free.

Other bits and pieces (that could have been points I talked about):
– get to know your fellow PhD students – you’re all in it together, experiencing the same highs and lows. The stats group have a weekly coffee event, not sure about the maths people.
– meet with your supervisors regularly even if you think you haven’t made much progress. I often find I make more progress in the hour or two before meeting with my supervisors than in the whole rest of the week!
– treat it like a 9-5 job so that you know you’re spending enough time on it.
– start writing early.

phd051412s

By | 2016-10-15T05:47:43+00:00 March 10th, 2014|Statistics|0 Comments

Quandl and R

logo_rgb_300
I haven’t taught econometrics for over a year now, but the next time I do, I’ll be using Quandl!  Quandl is a repository of data: “when a user clicks on a dataset on Quandl, the Quandl engine goes to the original publisher of that data, retrieves the freshest version of that data, and presents it to the user.”  There’s a lot of data online, but it’s really nice that they’ve aggregated so much here and made it extremely easy to access.

They have a nice R package that hooks into the Quandl API, allowing you to seamlessly (once you have your authentication token) import data direct from their servers – circumventing one of the major issues for new R users – importing and getting the data structured correctly.

They’ve provided an extremely brief “econometrics” tutorial and their own R cheat sheet.

 

By | 2016-10-15T05:47:43+00:00 November 7th, 2013|R, Teaching|0 Comments

Stats jobs for undergraduates

Update December 2015: I’ve made a new page dedicated to the various kinds of jobs people with statistical training can apply for (including job descriptions). Check it out here: garthtarr.com/jobs-for-statisticians

I regularly get asked for advice about what undergraduate stats majors can do after their degree (particularly if they don’t want to end up in a bank or consulting company). The standard response is that statisticians can do anything, but if you want to use your stats skills specifically, here are some resources:

Government

Lots of government departments take undergraduate and honours level statisticians, not just the ABS but also ATO, DEEWR, Defence (and specifically DSTO), ABARES, RBA, Treasury, Bureau of Crime Statistics & ResearchStatistics NZ … keep an eye out early in the year for grad programs. Also look into summer internships (e.g. ABS cadetships; RBA cadetships and the ABARES Summer Vacation program).

You could always become a teacher – not enough maths teachers at the high school level (or at the primary school level).  See, for example, the Teach for Australia program.

Private sector

Most (if not all) companies will appreciate a person with solid quantitative skills.  You could consider (to name just a few):

Within banks there are ways to use your statistics without doing financial work or trading.  For example the ANZ Bank has the Central Customer Analytics department and NAB has its Analytics and Research Operations department.

Further study

If you want to specialise further in statistics (without doing a PhD) you might consider a Masters in Statistics or Biostatistics. For example, UNSW has a decent Master of Statistics and the  School of Public Health here at the University of Sydney has a Master of Biostatistics.  There’s a program with NSW Health called NSW Biostatistical Officer Training Program which recruitments trainee biostatisticians every year (applications are usually due in November). While in the program, trainees work full-time in a variety of placements and undertake a Masters of Biostatistics part-time. NSW Health pays university and associated fees and study leave is given.  See also this blog post by Jerzy Wieczorek, mathematical statistician at the U.S. Census Bureau for some thoughts on Masters.

Job listings

You might want to subscribe to the ANZstat mailing list (make sure you set up a filter in your email program of choice so your inbox doesn’t get innundated with messages).  The jobs on this mailing list are often for people with a PhD but not always (for example, those NSW Health trainee biostats jobs get advertised on this mailing list).

There’s also the StatSci joblist and a page with more general information.

The Australian Mathematical Society (Aust MS) has a page on jobs for people with quantitative skills.

Sport statistics jobs

  1. Keep an eye on StatsJobs for potential openings. These are likely to be mostly higher level stats jobs (e.g. requiring a masters or higher) but there may be grad level positions. You could also keep an eye on the Sports Management Australia and New Zealand site.
  2. Go for positions in sports companies/relevant government agencies without a focus on stats, then (after a period of time) transfer into a more stats based job (if you go for a government job, they’re often really good about supporting further study, e.g. masters in stats). E.g. Department of Sports and Recreation 
  3. If you’re planning on heading overseas, the Royal Statistics Society (UK based organisation) has a Statistics in Sport section or the American Statistical Association have this advice. Unfortunately, there’s no equivalent in the Statistical Society of Australia Inc (SSAI).
  4. You could also look at companies like atass sports (UK based) or Statistical Sports Consulting (USA based). A dedicated stats company like this would give you the extra training in the appropriate areas that you’d need. But there doesn’t seem to be anything comparable in Australia (that I’ve been able to find). The next best would be to look for jobs with the Australian Institute of Sport, AFL, NRL, etc. directly.

City jobs

Most people know about the standard jobs in the city: investment banking, derivatives trading, management consulting, human resource consulting, other forms of consulting, … Those companies do a good job of getting the word out on campus about internships and grad positions.

By | 2016-10-15T05:47:46+00:00 August 28th, 2013|Statistics, Teaching|0 Comments

Recommended Reading

A student today asked if there were any books on statistics that I could recommend. He was after more generalist type books. I ended up sending him this list:

  1. The Theory That Would Not Die: How Bayes’ Rule Cracked the Enigma Code, Hunted Down Russian Submarines, and Emerged Triumphant from Two Centuries of Controversy (USYD, Amazon) is a great (generalist) read on the progression of bayesian statistics. It’s a really fun read (for a book about statistics).
  2. The lady tasting tea : how statistics revolutionized science in the twentieth century (USYD, Amazon) I quite enjoyed this one, it’s nicely written history of some key stats players.
  3. Statistics on the table : the history of statistical concepts and methods (USYD, Amazon) is a bit dryer than the above two – I haven’t made it all the way through yet – waiting for a rainy day!
  4. Mostly harmless econometrics : an empiricist’s companion (USYD, Amazon) is quite a bit more technical than the above books and more focussed on econometrics (statistics for economics).
  5. The Signal and the Noise: Why So Many Predictions Fail — but Some Don’t (Amazon) I’ve never read this but Nate Silver’s pretty hot right now.
  6. Probabilities : the little numbers that rule our lives (USYD, Amazon) I’m reading this on and off at the moment – it has some interesting observations.
By | 2016-10-15T05:47:47+00:00 April 11th, 2013|Statistics, Teaching|0 Comments

Australians love their cash!

Intrigued by this article in the SMH, I went and got some data from the RBA and the RBNZ. Using the googleVis package, available on CRAN, I made this chart to compare the value each person holds on average:

We can also translate this into the average number of notes each person has squirrelled away:

While it appears that Kiwis have more $20 notes per person than Aussies, it does seem that Australians have a stronger affinity for the higher denomination notes.[/fusion_builder_column][/fusion_builder_row][/fusion_builder_container]

By | 2016-10-15T05:47:47+00:00 February 14th, 2013|Feature, Statistics|0 Comments

The Elements of Statistical Learning

Also not really my field, but just found out that “The Elements of Statistical Learning: Data Mining, Inference, and Prediction” is available online for free. An extremely comprehensive text book written by giants in the field – it’s amazing that they were able to make it available for free (though if you find use for it you can (should?) still buy a hard copy).

Available here.

By | 2013-08-28T11:59:39+00:00 January 16th, 2013|Statistics|0 Comments

Odds Ratio and Relative Risk

I’m far from an epidemiologist, but odds ratios and relative risk come up often enough that it’s handy to have a solid understanding of what they mean.  These measures are used when faced with contingency tables:

begin{array}{c|cc|c}    & D^+ & D^- & text{Total} \ hline    S^+ & a & b & a+b \    S^- & c & d & c+d \ hline    text{Total} & a+c & b+d & a+b+c+d    end{array}

where D^+ is having a disease/condition/event under study and D^- is not having the disease/condition/event under study.  Also S^+ is testing positive/symptomatic/presence of a particular trait and S^- is testing negative/asymptomatic/not having a particular trait.

Odds ratio

The odds of success is the ratio of the probability of success p to the chance of failure 1-p:

text{Odds} = dfrac{p}{1-p} .

In the context of disease testing, we’d consider the odds of a disease for S^+ people (those with particular traits) against the S^- group (people without a particular trait).  The probability of having a disease for the S^+ group can be found by restricting attention to the S^+ row (restrict attention to the people who have the trait) and working out what proportion of those people have the disease:

P(D^+ | S^+) =dfrac{a}{a+b}

and the probability of having a disease for the S^- group is

P(D^+|S^-) = dfrac{c}{c+d}.

Hence the odds of disease for S^+ patients is:

text{Odds for }S^+ = dfrac{P(D^+ | S^+)}{1-P(D^+ | S^+)}

and the odds of disease for S^- people is:

text{Odds for }S^- = dfrac{P(D^+ | S^-)}{1-P(D^+ | S^-)} .

Finally the odds ratio is:

text{Odds ratio} = dfrac{text{Odds for }S^+}{text{Odds for }S^-}=dfrac{ad}{bc}.

That last step is just algebra.

What does it mean?

The odds ratio is a measure of effect size – how much of a difference does the positive test/symptoms/particular trait have on your chances of getting the disease?  An odds ratio of 1 indicates that the disease/condition/event under study is equally likely to occur in both groups (that is to say D and S are independent of one another). An odds ratio greater than 1 indicates that the disease more likely to occur in the S^+ group than the S^- group. Similarly, an odds ratio less than 1 indicates that the disease is less likely to occur in the S^+ group.

For example  an odds ratio of 2 indicates that people from the S^+ group had twice the risk of having the disease as people from the S^- group.

When can you use it?

The odds ratio can be used in observational studies (examining the effect of a risk factor/symptom on the disease outcome), prospective studies (where subjects who are initially identified as “disease-free” and classified by presence or absence of a risk factor are followed over time to see if they develop the disease) and retrospective studies (subjects are followed back in time to check for the presence or absence of the risk factor for each individual).

Relative risk

The relative risk is a measure of the influence of risk on disease.  It is the probability of contracting the disease given you have the risk factor divided by the probability of contracting the disease given you don’t have the risk factor:

text{Relative risk} = dfrac{P(D^+|S^+)}{P(D^+|S^-)} = dfrac{a/(a+b)}{c/(c+d)}.

What does it mean?

A relative risk of 1 means there is no difference in risk (of contracting the disease) between the two groups.  A relative greater than 1 means the disease is more likely to occur in the S^+ group than in the S^- group.  A relative risk less than 1 means the disease is more likely to occur in the S^- group than in the S^+ group.

For example a relative risk of 2 would mean that S^+ people would be twice as likely to contract the disease than people from the S^- group.

When can you use it?

Relative risk can only be used in prospective studies – note the wording above is all in terms of “contracting” the disease.  It is often used to compare the risk of developing a disease in people not receiving a new medical treatment (or receiving a placebo) versus people who are receiving an established treatment.

Odds ratio vs relative risk

Odds ratios and relative risks are interpreted in much the same way and if a and c are much less than b and d then the odds ratio will be almost the same as the relative risk.  In some sense the relative risk is a more intuitive measure of effect size.  Note that the choice is only for prospective studies were the distinction becomes important in cases of medium to high probabilities. If action A carries a risk of 99.9% and action B a risk of 99.0% then the relative risk is just over 1, while the odds associated with action A are more than 10 times higher than the odds with B.

This not being my area, naturally I turned to Wikipedia, which suggests that the odds ratio is commonly used for case-control studies, as odds, but not probabilities, are usually estimated whereas relative risk is used in randomized controlled trials and cohort studies.

Finally (and the real motivation for the post), an award winning video has been made by Susanna Cramb discussing the differences between odds ratios and relative risk:

By | 2016-10-15T05:47:47+00:00 January 10th, 2013|Statistics|0 Comments

Hans Rosling’s 200 Countries, 200 Years, 4 Minutes

Hans Rosling from The Joy of Stats on BBC Four. Another excellent example of data communication. I use it in first year lectures to elicit discussion on the issues with aggregating data, in particular how a summary statistic can hide differences between subgroups. We also talk about how many variables are being plotted. It’s something different for them – it puts what they’re learning in a global context and shows statistics as being more than just calculating means and variances.

Pretty neat, eh?

By | 2016-10-15T05:47:47+00:00 December 15th, 2012|Statistics, Teaching|0 Comments