10 essential skills: (1) applied statistics, matrix theory, and mathematical reasoning

My story: I dropped out of "maths" early in high school. And if you think about this for a moment, it means that by the time I could legally drink alcohol, I still couldn't solve a basic algebra equation for the single unknown "x", to save my life.

Side bar: In Canada, a student can elect to forgo further studies in mathematics half way through high school. I think this is a terrible shame. The opt-out rate - in my opinion - is a reflection more so on the structure of the curriculum and teaching methods, than the failure of students to internalize mathematics' "riveting" universal truths. I maintain that proficiency in mathematics takes practice, perseverance and patience. As such, it deserves to be appreciated and considered carefully by young people before ignoring it almost entirely. After all, it may just come back to haunt a person... read further.

My story, continued: When I started my university studies in 2008, it was the peak of the financial crisis. I witnessed the prospects of work within my "dream" profession dwindle around me - journalism was dying. Or so it appeared. Major newspapers were laying off their writers, shutting down, or consolidating to keep afloat. Ironically, I felt so overwhelmed being unable to explain, let alone comprehend, the events unfolding around me that I decided I needed to study finance and economics!

At 19 I spent my summer away from university to sit - full-time - in a secondary school classroom studying advanced functions and calculus with teenagers. The reason for this is that a university will audit your high school transcript to ensure you meet the basic requirements for a sciences degree, thereby necessitating my impromptu return. (Take note of this please, young ones). And so, at 19, a very humbled me sat in a very warm classroom for the months of July/August studying mathematics.

Fast forward four years: I've completed my Bachelor of Science with Honours in Economics and at least 10 of the 20 credits on the transcript were courses in introductory/advanced statistics, mathematics, and algebra.

Conclusion: This is how a 16 year old imagining a life of investigative journalism and novel-writing dropped out of high school mathematics but today describes the journey of data science and machine learning engineering.

The lesson: don't skip the mathematics. In fact, try not to give up on anything you can't immediately understand or solve without giving it the time and effort.

***

So you're thinking about a profession in data science or the data analytics profession more generally?

Step 1: Foundational knowledge in applied statistics, matrix theory, and mathematical reasoning.

I do believe a STEM degree is necessary to pivot into data science, data mining, machine learning, and all other related roles in this burgeoning field. The acronym STEM presents the disciplines of Science, Technology, Engineering, and Mathematics. I believe these seemingly unrelated disciplines are all connected at heart by the graduate's foundational knowledge of applied statistics, matrix theory, and mathematical reasoning.

I advocate in favour of a STEM education because I know just how much my education has shaped my critical thinking skills and made me a self-learner in the technical domain. Importantly, it has also given me the confidence and tools to know how to solve the problems I'm sure to encounter in the professional world.

And yet, I have met many people working at the heart of technology who haven't completed a university education, let alone, attained today's conventional standard of a master's degree in their related field.

Irrespective of your academic endeavours, if you are truly looking to lead a professional career in big data, then a responsible effort to better understanding big data's underlying technologies is due. That includes:

• Developing a proficiency in applied statistics

Not tonight, not even tomorrow, but everyday moving forward. Baby steps. Here are some resources to get you started:

Free online textbooks:

Introductory Statistics PDF - this is a fantastic online textbook that can be downloaded for free from Rice University. It is well organized and covers important topics in introductory statistics.

Statistics and Data Analysis PDF - this is the online textbook for introductory statistics at Western Michigan University and is written by A. Adebe, J. Daniels, J.W. McKean, and J.A. Kapenga. It covers basic descriptive statistical and graphical procedures for analyzing data sets with some exposure to simple inferential problems to parametric and non-parametric procedure. It is an easy to read and concise overview of the very basics.

Low-cost online courses:

More simple:

Become a Probability and Statistics Master - this Udemy course includes 360 practice questions to test your knowledge and currently costs $24.99 USD.

Expensive and intermediate:

MITxPro Statistics and Data Science MicroMasters

Here is a great short article on Towards Data Science by user George Saif "The 5 Basic Statistics Concepts Data Scientists Need to Know." It's a wonderful introduction to applied statistical concepts!

• Knowledge of matrix theory and linear algebra

Free online:

Introduction to Matrices PDF - this is a fantastic review of matrices, and a quick read. It will help introduce you to the matrix logic behind big data, data science, data architecture, etc.

Low-cost online courses:

Data Science Linear Algebra Complete Guide - Machine Learning - this Udemy course emphasizes the mathematical intuition required to work in the field of data science and machine learning and currently costs $24.99 USD.

Mathematics for Machine Learning - Linear Algebra - this is a Coursera course that is part of the Mathematics for Machine Learning Specialization. You can even consider this specialization if you'd like to follow a structured curriculum online and are motivated by the certificate of completion! There are three courses in the specialization: multivariate calculus, linear algebra, and Principal Component Analysis.

Here is a great article on TowardDataScience that describes the key takeaways from linear algebra. It's written by the user hadrienj and I loved its simple visual demonstration of the concepts.

***

The "best" way to get started is to be patient in the process. Mathematics is a broad topic. Sometimes, the foundational logic essential to understanding and completing basic exercises can take some time for a person who hasn't studied the material in a while. That being said, the sooner you start, the better!