Nowadays, data science has been a hot topic in the banking industry. Big data, Business Intelligence, Block Chain… new technology has gradually reframed the bank structure. Professionals with excellent data analysis skills are always in demand for the job market. Following the trend, more and more candidates become interested in application of data science in work environment. NHC DSS held a Data Science Talk on Saturday, May 13th, which covered a SAS Tutorial and a Case Study in Market Analysis.
Why SAS (Jack)
At the beginning of the lecture, Jack, from RBC CIDM Group, started with a big picture of data analysis area and talked about core skills required by the industry. He pointed out the importance of SAS in a data analysis related work. As an experienced interviewer, he also showed us some typical interview questions for data analysis jobs.
SAS Tutorial (Sam)
In the first section, Sam, Founder of SASCrunch, gave us a hand-on tutorial on SAS manipulation. Sam started with code structure and common errors of SAS. He then introduced us some important concepts, such as temporary & permanent libraries, data sets and six attributes of a data variable. In practice, data scientists need to communicate well with client-side on data sets before doing any analysis work. Therefore, Sam showed us an efficient code to display data overview. For the next step, he mentioned some basic codes for data set, including data set creation and selection. In order to guarantee efficiency and consistency of large dataset analysis work, data analysts need to create a folder, called data library, to save all their work. In this way, redundant work can be avoided by creating a permanent library under the path and saving all work into the library. In a nutshell, Sam’s tutorial shed some light on application of SAS in real work and how to start SAS learning.
Open Sesame! – A brief introduction on Marketing Analytics & Product Design in Retail Banking industry (Moiz/Audrey)
In the second section, Audrey and Moiz from Scotiabank Marketing & Segmentation Team gave us a brief introduction of career development on marketing analytics side and walked us through a typical lifecycle of lending product analysis process.
Audrey shared with us her opinion on what makes a competent candidate in the marketing analytics teams. She mentioned fair technical background with SAS/SQL/Tableau, understanding of finance knowledges, excellent negotiation skills and business acumen as four key factors to succeed in this industry. She also showed us some possible career paths in banks based on the above mentioned skillsets, such as job opportunities related to Products & Marketing, Customer Insights, Risk Control and Business Intelligence. Instead of focusing on technical roles which used to be popular among Chinese professionals, Audrey gave us an insight into how to choose or shift to a marketing or finance consulting career path based on our own background and interest.
Moiz’s speech focused on using business case with SAS to explore product database and marketing design process. First, he summarized personal banking products as four categories, including day-to-day deposit products, term deposits, term loans, and revolving lending products. He then showed us the End-to-end loan life-cycle analysis & analytics process implemented by retail banking industry and emphasized on key skills required for each stage. In order to carry out proper marketing or management strategy, the database marketing team will typically go through four stages to figure out answers about “sell to whom?”, “What/when to sell?”, “How to re-sell?”, and post-analysis retrospective. He mentioned that major data manipulation work is done by SAS, and Tableau is mainly used for data visualization or fast clustering process.
On top of that, he presented a hand-on demo of Mortgage Offer design to us. The demo is based on a Customer Preference Survey conducted by a consulting firm. According to data collected by the survey, customers are clustered as three groups with different interest rates and service level expectations of a mortgage product. He also introduced typical P&L analysis based on cost of fund and expected loss of the product. After the financial analysis, the marketing analytics team used yield model to design proper offers to cater the 3 customer groups. Technical skills were used in this case include SAL, SQL, Excel and Macro. This STP (segmenting-targeting-positioning) marketing approach based on large volume data helped us have a rough idea on how the analytics consulting role is working in banking industry nowadays.