mini传媒

Published on July 5, 2024
Job titles and other details are as of the time of publication.
(The interview was conducted in Japanese and was thereafter translated into English.)

Three research fields that handle huge amounts of data

My research has three major pillars: “missing data analysis,” “marketing science,” and “empirical accounting.”

The first is the field of “missing data analysis” using Bayesian statistics. In consumer surveys, for example, there are cases in which respondents hide information such as their place of employment or annual income. Such data cannot be used for corporate marketing as it is. Therefore, we use the theory of Bayesian statistics to predict and fill in the missing data with other data. Even in cases that are difficult using traditional statistics, Bayesian statistics can lead to more accurate inferences.

Next, “marketing science” is a field that approaches consumer behavior from data such as location information, purchase history, and in-store movement information. Previously, I was consulted by a soft drink manufacturer to examine the impact of flyers for “High & Low” supermarkets, where cheaper products change daily, on “Every Day Low Price” supermarkets, where product prices do not change daily. I obtained and analyzed flyer data, smartphone GPS data, store receipts, etc. from 80 stores in 10 regions, and found that there is little point in selling products that damage the brand value at “Every Day Low Price” supermarkets. I submitted the research results so that the manufacturer could make use of them in their sales strategy for retail stores such as supermarkets.

Finally, “empirical accounting” is a field that uses the financial statements of listed companies to study how accounting system reforms affect companies and how they relate to stock prices. What I am particularly interested in is “the relationship between the size of an audit firm and the quality of its audits.” In this study I used causal inference, a statistical method, and obtained the result that “differences in the size of audit firms do not affect audit quality.” Recently, I have also been conducting research to explore the extent to which AI can replace the work of certified public accountants.

All of these studies handle huge amounts of data, but I engage in daily research with the idea that it will lead to solutions to social and business issues, rather than completing a study as simply a study.

Learning accounting at a vocational school while attending university

I have always loved mathematics. Since I was enrolled in a liberal arts course from my second year of high school, I vaguely thought that I would probably enroll in the economics department of a university. I entered the economics department to study accounting, with the goal of obtaining a certified public accountant qualification for my future. However, since I was planning to study not only at university but also at an accounting vocational school, I considered the cost and chose Nagoya University, which I could attend from my parents’ home in Gifu. I commuted to university from my parents’ house every day.

I also liked English in the sense that you could solve problems if you understood the rules. It may be difficult to memorize the vocabulary, but once you overcome that problem, the rest is according to the rules. At vocational school, I studied for the USCPA (United States Certified Public Accountant) and BATIC (international accounting test) in order to improve my abilities in both mathematics and English. Speaking of schools, it was also around this time that I started going to a golf school. I played baseball all the way up until middle school and loved hitting the ball. But I wasn’t cut out for team sports, so I replaced my bat with a club and started playing golf. Perhaps because a sport that can be played individually suits me better, I still continue to play golf.

Discovering the excitement of statistics and participating in a master’s course accounting seminar as an undergraduate

During the first half of my university years, I continued to study accounting by attending bookkeeping classes. When you learn to read financial statements, you can get a clearer idea about a company. I found that interesting, and the knowledge I gained was very useful as I am still doing research in accounting. On the other hand, I thought that studying to become a certified public accountant did not suit me. There were two reasons for that. First, if you want to become a certified public accountant, you will need to study subjects that require memorization, such as the Companies Act and the Financial Instruments and Exchange Act. Just like with English words, I was not good at memorization.

The other reason was that I became seriously interested in statistics, which I had been studying since my first year. By using complex methods such as statistical analysis, you can get a quick idea of a company, similar to reading financial statements in accounting. Another thing that attracted me to statistics was that I could make use of mathematics, which I was good at. Therefore, I took advantage of the university system and participated in a master’s course accounting seminar in my fourth year, and in addition to accounting, I also deepened my learning about statistics.

Collaboration with companies that provide data is essential

While I was taking part in the accounting seminar, I had developed the desire to do research that tied together business and academia, but I was also hunting for a job. I received job offers from two private marketing research companies and was planning to work as a data scientist. However, my supervisor advised me, “If you want to tie the two together, why not try researching many companies in the academic world instead of working at a specific company?” This led me to pursue a master’s degree and choose a marketing science seminar. However, I remember that there was little joint research with companies at that time. Unlike today, when we are constantly receiving offers for joint research, the field of marketing science was probably not yet recognized by companies.

However, my desire to do research that tie together business and academia did not change. In marketing activities using data, companies often provide the data. It is also true that research themes can often be found within that data. In that sense, I did not have the option of completing my projects in the academic world without collaborating with corporations.

Expectations companies have for the Faculty and Graduate School of Social Data Science

In this way, I was particular about working in a space between business and academics. I had no hesitation at all about participating in the launch of the Faculty and Graduate School of Social Data Science (SDS), and immediately replied, “I’ll join.” Currently, I am also in charge of public relations activities. For example, the Hitotsubashi University SDS exhibited a booth at a recent event held at Tokyo Big Sight (Marketing Technology Fair Tokyo 2024). It was more fun than I expected to give lectures and exchange business cards with people from companies.

Two things I realized from talking with people at companies are that there is a great need for joint research, and that they are interested in SDS master’s program students. For example, I learned that graduate students are expected to be leaders in joint research between SDS and companies. Compared to when I was enrolled in my master’s program, it has become much easier to go back and forth between business and academics, which I find it very good.

The ideal is for students to act as a hub and connect researchers from different fields

I myself hope that students will serve as hubs that connect us researchers. For example, when I supervise a master’s student’s thesis and I become the supervisor, I will ask a professor in a completely different field to serve as my co-supervisor. By supervising the student, the professor and I will collaborate, compile the student’s paper, and submit it to an academic journal. I think that’s the best way to do it, and I have a feeling that this will come true because SDS researchers are very close to each other and have active exchanges.

If you are a high school student who would like to use your math skills to help solve social issues, but don’t know exactly how to go about it, SDS may be the perfect environment for you. Researchers from a variety of specialized fields gather here, including economics, business administration, law, psychology, the environment, AI, real estate, and sports. It is not too late if you decide what field you want to use your mathematics knowledge in after joining SDS. SDS is an environment where students who are eager to use data to solve social issues gather and learn before moving into the real world.