A Data Science Central Community
Hi all, hope someone may be able to give me some thoughts on career paths and education. What is this group's opinions on which of the two above courses gives one a better background to practise data science?
I've just completed a double bachelor of science and arts, with majors in psychology, neuroscience, history and philosophy. I think that my ability to practise now would be strong, as I have good experience with real life psychological and neuroscience datasets, and a research background far stronger than most undergrad's (expect to co-author a conference talk to be presented 2013). Through independent learning I've gained a good understanding of linear algebra, databases, machine learning and stochastic processes, however I find it hard to communicate this to prospective employers as these are not formal qualifications.
When I approach the market I'm consistently offered lower pay for grad positions than I'm currently receiving in my mid level administrative role, and would reasonably expect in other such roles. I'm thus trying to piece together a formal qualification in finance, traditional statistics and machine learning so that I can gain the best breadth of exposure to all aspects of data science, however the courses offered here in Sydney are split into computer science and statistics courses, with no universities offering a course which encompasses aspects of both.
As a result, I'm basically which of these qualifications will give me the better chances of landing a good job long term, while still teaching me something I don't actually know. My mathematical background is probably my weakness at the moment, which is why I'm leaning towards the MQF; it also may help me stand out in a sea of MBAs with IT backgrounds.
What are the thoughts of the group?
We are working on a Proposal for an Apprenticeship in Data Science. This "course" would cover both aspects and even more in a very short time period (< 6 months).
I think this is a great idea for professionals already working in a related field whose employers might cover the cost. I wish I had the capital to outlay on an online course at the moment; I'm also familiar with some of that material currently, although I'm sure I'd still gain a new perspective on it.
The way things stand in Australia at the moment a local degree is still seen as more credible in the eyes of employers; unfortunately many here have a closed mind when it comes to online education, and it's (unfairly) seen as a second rate qualification. This is obviously ridiculous as I can attest that plenty of online education (Stanford, MIT especially) I've undertaken outclasses the on-campus courses I've undertaken at one of our top uni's!
This was what I was thinking also. I think the majority of people in middle management now have an MBA - uni's in Sydney are now offering "Executive MBAs" to help real leaders stand out because MBAs have become so commonplace.
I was going to treat the MBA as a back door into a masters in IT really, but I think this material might be easier learned independently than the mathematical material from the MQF.
A MS in Statistics is also coming up as an option now... Probably doesn't sound as impressive as the MQF, but it does have a wider variety of courses in optimisation, data management, data mining as well as the ability to take courses from the school of computer science to provide full coverage of both statistical and computational fields.
I'm not sure if I can get entry to the course with my background however - psychology and neuroscience sometimes aren't considered quantitative fields. Despite that challenge, maybe this is the best option for me; I'm just hoping it isn't viewed as too general by employers.
Essentially, it sounds like you are trying to go down the Operations Research route. If you are weaker in math, perhaps getting some high level statistics courses under your belt will help. I find it hard to believe that people are having difficulty understanding some of your knowledge base. I think part of it is that you need to use common language. Find some job descriptions for manager of analytics, operations research manager, operations analyst, etc and look for the keywords they are using to describe the needed skills. You have many of them.
If you have a good understanding of relational databases and Oracle, SQL, etc, and used SPSS/SAS/R (statistical scripting languages) those HAVE to go on the resume. Recruiters understand those terms and are very common in JD's here in the states. If you talk about stochastic processes, you will lose them. Leave it at mathematical or quantitative modeling. A good example of this is the application of linear algebra (or linear programming) will probably be optimization of networks or processes using constrained resources. Someone in HR will not know what linear algebra means, but they do know "optimize."
More importantly, provide specific examples of your book knowledge that you used in a practical situation if you can, but leave the jargon for the interview with the actual hiring manager that knows what those terms are because they use those methods themselves. Good luck!
Thanks for your reply Robert - I think you have some good points here. Incidentally, my interests are operational, technological as well as financial, I'm interested in machine learning/recommender systems, advertising, CRM, finance and risk. Basically anything involving modelling of human behaviour (see that psych background coming through!)
I do think that there is a tendancy to just look at the title of someone's degree sometimes and just write them off if it doesn't specifically say maths or stats - guessing this is why recruiters are struggling to understand my background, because I'm not your typical psych graduate. I'm familiar with SQL, SPSS and R, as well as having some knowledge of C# and python. I would have thought (like you) that these would get you over the line when talking to recruiters, but to be honest I think I've received better responses when talking more about my soft skills.
I guess this comes down to (like you said) couching things in terms recruiters understand, however I think sometimes (as Ed mentioned) there is a disconnect in terms of pleasing the HR people with soft skills vs pleasing the hiring manager by being technical. It's hard to do both, and if you get through the recruiters who act as gatekeepers, you then have to pivot to capture the interest of the hiring manager.
Unfortunately you are encountering a phenomenon occurring in many sectors where the producers are very tech-centric but the C-Suites aren't investing sufficiently in HR personal who are tech savvy enough to effectively fill the positions for which they are responsible. Target your search and language to communicate with the hiring manager (find them via linkedin) and have them march you down to HR and say "hire them".
I think this answer really gets to the core of the issue here! I've been told by recruiters before that they "aren't technical" and "don't understand" the technical details; basically they have no idea what the candidates would actually be doing in the role. As a result it's hard for me to show them that I can deliver more value than the average graduate.
Unfortunately, a lot of positions are advertised only through recruiters, who usually hide the identity of the client, so it's hard to track down the hiring managers. That said, I think doing some research through Linkedin to target analytics focused businesses to see if there may be opportunities directly may be a good option which I should probably take further advantage of.
Is the masters of Quant Finance an engineering degree?