Tell us about yourself
I intend to ruminate on my experiences in the last nine months in the graduate program in Business Analytics (MSBA) at the Carlson School of Management in the University of Minnesota. I hope this post would shed some light on the MSBA program, which is relatively a new program in the U.S, and thus benefit prospective students and aspiring data science professionals.
Original Link :
How did you end up in such an offbeat, unconventional and unique career?
I assume most of the people reading this post know why there is a need now for a new program like the M.S in Business Analytics. The reasons are multi-fold. The media, the entrepreneurs, the CEOs, and the data gurus have regurgitated the value of Big Data in this new digital world ad nauseam. Every Fortune 500 company in this modern world is actively leveraging the data they collect or buy to drive business decisions. Google, Amazon, Facebook, and Uber are but a few companies which are using cutting edge technologies to innovate and drive business processes. The motivation behind writing this post is not to talk about the “why”, as I believe there is a general consensus on this issue, but to talk about how this program trains new talent to fill the gap between demand and supply in the Data Sciences & Analytics industry.
Tell us about the programme
I firmly believe that the MSBA program at the Carlson School of Management has the one of the best MSBA curriculum’s and faculty in the U.S. It has everything it takes to start one’s career as a “rounded” data science professional in a top notch analytics firm. This 45- credit, 11-month highly intense program starts off with foundational courses in Statistics, Python, and Marketing Management. Marketing Analytics is a very good exit option post graduation and the Marketing Management course teaches students how data analytics is being used for decision making in the marketing industry. Besides the above three courses, the summer term also has a dedicated course to motivate and discuss various analytics case studies licensed from Harvard, Stanford, and Kellogg Business schools.
Come Fall semester, the program gets very intense with courses in Exploratory Data Analysis & Visualization, Predictive Modeling, Harvesting Big Data, Databases & Data Warehousing, and Analytics Project Management. These courses are taught by the best professors in Information Systems and Decision Sciences Department – Professor Ravi Bapna, Professor Gedas Adomavicius, and Professor De Liu. This was a period of immense learning and I remember not having slept for more than seven hours on any day in this semester. Few of the concepts learnt in this semester are:
1)Various Data Cleaning ,Data Reduction & missing value imputation techniques , Principal Component Analysis, Advanced Clustering techniques ( DBSCAN, Probabilistic Clustering, PAM), Topic Modelling(LDA), R Shiny, and Tableau.
2) Predictive modelling techniques such as Naïve Bayes, KNN, Logistic Regression, Decision Trees, Random Forests, Support Vector Machines, and Neural Nets. Bagging, Boosting, Stacking (Ensemble methods), Meta Modelling, Text Mining, Lift Charts & ROC curves, cross-validation, Personalization & Recommendation Algorithms.
3) Harvesting Social Media data (Twitter and Facebook) using Web Scraping and APIs, Hands-on learning of Big Data technologies such as Hadoop, Hive, Pig, Mahout, and Apache Spark. Usage of Amazon Web Services (AWS) to set up and work on Hadoop clusters
4) Advanced SQL, Data Modelling and ER diagrams, Data Warehousing concepts, ETL concepts using Microsoft SQL Server & Visual Studio and NoSQL databases (MongoDB)
5) Managing analytic projects using Agile, Traditional Waterfall, and Kanban methodologies, managing team & client relationships, defining project scope and timelines, and writing project charters and statements of work.
Are there any real application based projects to test theoretical understanding?
Apart from the weekly assignments, each of the courses had a group project towards the semester completion which gave us an opportunity to apply all the concepts learnt thus far on real world business problems. By the end of this semester, I could clearly see a transformation in my approach to solving business problems. To re-emphasize, each of the courses is taught with the business use cases in mind and everything we learnt in each course had a clear connect to the real world problems. Python and R are the two most commonly used analytic tools in the industry and hence most of the applications of the concepts we learnt in the class are applied using these tools. However, for predictive modelling we used RapidMiner to an extent.
After a 20-day fall semester break, we started our final semester. In this semester we took three courses and each of us are put on a 14-week, semester long Experiential Learning Project (ELP). In the ELP, teams of five students work with clients from Fortune 500 companies to solve business problems which are currently a concern to them. Each team typically puts in anywhere from 15 to 30 hours per week on these projects. My team and I are currently working with a major global strategic consulting firm and are designing personalized ranking algorithms (like PageRank) to reduce search frictions and search time and hence improve the information retrieval process in their enterprise search engine.
Coming to the coursework, we are currently learning Time Series Analysis & Forecasting using ARMA/ ARIMA models , Stochastic modelling, and Data Structures & Algorithms ( Minimum Spanning Tree, Max flow/Min Cut , Shortest path) in one of the courses . Causal Inference, AdFX effects, AB testing, Propensity score matching, and experiment design are being taught in the other course. We are yet to start one other course which deals with Linear Programming, Data simulation and Risk Analysis.
Final word to students?
I am sure that after reading this post you would see the value in joining this cohort. In the past three to four months, I have been talking to many of my juniors from BITS Pilani, Ex colleagues from Mu Sigma and other people who are considering a career in Data Sciences and Business Analytics.This post is a honest attempt to provide an insider opinion to mellow the fears and concerns of prospective international students who are planning to take up this course.
The last batch witnessed 100% Placements with few students receiving multiple offers^1. I was offered a position as an Analytics consultant in a California based firm but decided not to accept it due to various reasons.
In summary I conclude that the future is bright if you make the right decision! Thanks for reading through.
Srinivas currently works at Mckinsey & Co as an analytics fellow. He is part of the advanced analytics team – leveraging business, statistics, and analytics capabilities to solve complex business problems.