When you study MSQE course at ISI, you are trained with a significant amount of
mathematical and statistical skills and tools for doing economic research.
And, when you read econometrics or linear programming or let’s say you study statistics, or optimization, you have a primary objective to understand, “how do I do the economic research or economic analysis with these tools”, and “are these tools helpful for me or not”.
Now having said that, it does not mean that you have to restrict yourself to economic research. The idea here is that economics is a lot about model building and number crunching. If you try to understand it, data science is kind of pretty similar to economics and the skills are transferable to each other. So, you can transfer your learnings in MSQE as an economics student to a data scientist and rather I would say, a data professional.
I myself had qualified for both, CMI’s Master’s in Data Science program and ISI Kolkata’s MSQE program. Now, I chose MSQE program at ISI Kolkata mainly because of few reasons. I also wanted to study a bit of economics and finance. One of the things about finance is that, if you do not understand economics properly, then your finance knowledge will be like the people who come to news studios or business new channels and they keep on talking about things without making a lot of sense. So, you have to understand economics and not just the macroeconomics but the microeconomics parts as well. This is so because without a firm understanding of microeconomics, your modelling skills will suffer a lot as only when you study microeconomics properly, you try to understand the understand the significance of economic modelling in situations which involve transactions and incentives.
When you move on from microeconomics to macroeconomics, there are a lot of factors which are governed by your set of assumptions and in the case of data scientists, things are slightly different.
Data scientists are people who are more akin to throwing a lot of computation power and a lot of algorithms to data and they do not really bother about the assumptions in lot of cases. This could be controversial for a lot of people, but it’s pretty accurate that when you are a macroeconomist, your assumptions are like fundamental basis of your every analysis. If I change one of the assumptions of your model or your analysis, your whole analysis will change and that is something which is not really so much apparent in microeconomics or any domain of natural sciences.
So, that is where the departure starts happening when you move from microeconomics to macroeconomics. But that is also the part where data crunching begins when you move from microeconomics to macroeconomics. In macroeconomics, you access GDP data, inflation data and try to make sense of what will happen with these numbers and how will exchange rate, trade surplus or every other macroeconomic factors will affect these numbers.
If I have to give you a basic idea that what is really a very prevalent theme in your
macroeconomic research, its understanding of time series analysis. Now, when we talk
about time series analysis, they are essentially two types of time series. One is univariate time series, in which you just have one unit and for that unit, you are observing a particular feature at difference points of time and just noting it down.
There could also be a time series in which you have multiple units and for all those units, you are observing a feature and noting it down at all time intervals. This time series becomes a panel data time series and earlier one was a univariate time series. If you go into the deepest part of econometrics and time series, you will have to deal with a lot of panel data. This is the part which is not really touched by data scientists but it is touched by econometricians and macroeconomists.
There could also be a time series in which you have multiple units and for all those units, you are observing a feature and noting it down at all time intervals. This time series becomes a panel data time series and earlier one was a univariate time series. If you go into the deepest part of econometrics and time series, you will have to deal with a lot of panel data. This is the part which is not really touched by data scientists but it is touched by econometricians and macroeconomists.
And, the time series forecasting or analysis where you calculate the trend seasonality or do the forecasting, is actually dealt by a lot of data scientists as well.
The problem here is that, a lot of people think that you need a lot of coding expertise in data science. That is true to an extent and it depends on the company and on the role. A lot of companies require software engineers and as they want to market themselves well, they label it as a data scientist role hiding it beneath a lot of jargons. So, you need to be sure that what kind of role you are getting into, rather you have to understand whether this is a software engineering which you are applying to or really a data scientist role. Not all companies will be very honest and upfront but all good companies are pretty much upfront what they expect and require from you.
Whatever you study in MSQE, which is a quantitative economics course, a lot of things are related to number crunching, data analysis, mathematical and statistical skills and all is for economics research and you can very well use it to become a data scientist.
As I have already told that I wanted to become a data scientist but I had a keen interest in the markets and finance, that is why I chose the MSQE program.
If you have any doubt and queries regarding this topic, feel free to comment down below.
