Introduction
I started my career as a software engineer and had a great time building applications. Those were simpler times, I could see what was built. Adding a button that appeared on the webpage and all one had to do was make sure the functionality works, data read /writes happened asynchronously, and the page can be scaled up to the user base it was intended for. Right when everything was going fine my interest in Data had piqued which is when I had decided to branch off to understand data and help organizations make best out of the it.
Masters
After months of research, GRE, Toefl, talking to students from the university it was decided, and I was set to start MS in Data science. Home sickness was definitely a part but the excitement of going back to school was overwhelming (I know, what was I thinking?).
We all looked like a kid high on candy.Yes, you guessed it right, all good things came to an end after a month into coursework. A combination of factors kept us going including getting accustomed to the campus, finding our way around, weather, making it to classes on time etc. Here is how a typical day would look in the life of a graduate's life.
- 6 AM - Get up and going for a campus job to earn your living
- 9 AM - Go for classes until 5 pm
- 6 PM - Watch lectures if required and work on assignments
- 8 PM - Catch up on outside course stuff, new technologies
- 9 PM - Yes, food. Cook, clean and eat to survive
Most of our days went by this way and with coursework piling up and the pressure to be industry competent was getting to us. With months passing by our coursework & advisers had told us that to be a good data scientist one had to know/ be comfortable in statistics.
Entry statistics
The following semester most of us ended up choosing statistics elective and the most shocking part was the concepts were Greek and Latin. From the minimal knowledge of hypothesis testing to knowing different ways of performing these tests (one way t tests to ANOVA), we all had to upgrade our learning skills quick enough. There was no room for one to understand why all this happened but just how this happened. It was essential for us to know them in order to apply them in ML methodology.
To be a good enough Data scientist one had to know decent statistics (even though some claim this may not be necessary). I believe one needs decent stats skills in order to create good models to solve Data problems in the industry.
Conclusion
Today most of us are successful Data scientists but we wish we had enough guidance to prepare us for the coursework we were taking. If an Alumni or mentor had mentioned to us in advance on how important statistics was for us to excel in Data Science we could have avoided the last minute struggle. Some of us would have had the time to understand how the concepts applied on real life scenarios to prep us for an advanced stat course.Being able to appreciate statistics as a Data Science professional is very important and comes in handy when performing experiments and while creating models that your organization could make use of. I still have days in my life where I go back and study up stat concepts in order to solve DS problems effectively.By: Avanthika Sankararaman