FREQUENTLY ASKED QUESTIONS
A set of ten questions, compiled by the founder, makes an attempt at adressing relvant issues and concerns that you - both as a parent and a student - might have as you make your mind to enroll. Data sciences is vast and an attempt, through this section, is just to simplify the doubts and confusion into words that addresses those concerns as best as possible. I have recorded videos for the first two question becasue I thnk the first two questions are highly relevant and, therefore. If still, there are any doubt, concern, issue that has not been addressed then you can drop me a whattsapp on my personal number
The world is fast moving towards automation and there is an increasing recognition of this at the macro level across both advanced economies and emerging market economies. Jobs would be lost but it essentially means that you need to take on the skills like data science skills would provide you the skills to not only take the jobs but also do well on those skills. The other thing to recognize is that much of what is in the modules is linked with academic research. Our economy is going to expand rapidly and with the advent of 5G that adbe the fastest growing this decade, there will be a huge demand for those who have the required skill sets. These skills would obviously also enable you acquire critical thinking skills that will
WHAT IS DATA SCIENCES?
The figure, given above, captures the essence of how knowlege of three distinct fields of study make up what is known as data science(s). Data sciences is a buzzword, for sure, and the new education policy has most certainly given it the limelight becasue technological innovation changes taking place in the world is is made up three elements. But, before, I come specficially to those three elements, let us take a moment to know what is the meaning of science. I took the meaning, verbatim, from vocublary.com, "Science is the field of study concerned with discovering and describing the world around us by observing and experimenting." The word obersving and experimenting are the key here. Data, of any kind, is firslty oberved. There are two kinds of data and they are structured data and unstructeed data. Anyhow, since this is not the point of this question so I would not go into too much details on data. Coming back to The experiment of it comes on later. Now, let us come to the three elements, below, I explain them.
The first element, and I would say the foundation, of it is the combination of mathematics and statistics. The knowledge of mathematics and statistics together make up what is known as economterics. Now, economsterics, as a discpline, was conceived in the early 19th and the term was coined back then. Many undergraduate students of econoics would be familiar with what economterics is about and many would not have even heard of this term. Economterics uses the foundations of mathematics and statistics to be able to design and build models in order to test theories or examine the linkages that might or might not exist. One of the most importnat parts of an econometrician is to underpin and identify causal relationship which is of the most fundmanetal problem in all of data sciences. Understanding the mathematical background is tke key and what helps in laying a strong foundation
The second is the knowledge of the computer science/IT. In here, I would tend to make the argument that you do not need to be a computer scientist or a programmer. These are nothing but the software packages that are used in order to implement the models that you have conceived of by brininging your critical thinking by examining the depth and breadth of the problem in hand. Most software packages have in built in them most of the data science toolkit and the models too. It is something that you will see during the lectures as the dummy models are implemented. There are some programming languages, like R, Python which require programming but even here most of the codes are readily available, for instance, most of the packages can be found in Github and those need to be modified in order to run the specific models for the tasks. Still, there are other software that do not require coding and are a manual. The point is that you can also learn to program and code yourself but this knowledge is not going to set you apart from your peers. Initially, the use of these softwares might be challenging but gradually as you would get used to using them would enhance your understanding of how to use them. I must point out that coding is one part of what constitutes data sciences. The essence is knowing the repertorie of the all the toolkit of the
The third is the business knowledge or the domain knowledge. More on what is business or domain knowledge has been explained in the F&Q 5. Therefore, I leave much of the explanation of the business domain knowledge to that specific question. One point that I would make here is that the business domain knowlege is something that you only pick up when you are on the ground. You can, nontheless, read about the five broad categories of business domain that are out there. For each of the business domain knowledge, I have certian notes and those I have uploaded in the dropbox. You could, if you are interested, acquire more than a fundamental knowledge of all the business domains that exists. Business landscape keep changing so the
Get to Know Us
Use this space to tell users more about yourself or to describe what your business does. Click to edit the text.
Watch this video if find the explanation boring.
WHY THE NEW EDUCATION POLICY HAS INTRODUCED DATA SCIENCES AT THE SCHOOL LEVEL?
Watch this video if find the explanation boring.
In fact, a quote for Howitt and Blake (2003) very aptly sums this situation up – ‘Where there is a child there is curiosity and where there is curiosity there is science.’
Now this is a very important question. Over the last hundred years, technology has rocked the world. So, there is now an increasing recogition of the fact that jobs are becomg more technical than ever. The demand for jobs which require this skill set will only go
‘Where there is a child there is curiosity and where there is curiosity there is science’ (Howitt & Blake, 2010, p. 3).
The pandemic has only ramped up the process of technology adoption so it is ever
Therefore, acquiring these skills are very importnat in order to have a decent change in the job market to be able to make a solid impression
As an economist, I would tend to make the case that India's.
The excerpt, given below, has been taken from the New Education Policy 2020 prepared by the Ministry of Human Resource Development, Government of India. This exceprt drives home the point as to why is this importnat.
"Indeed, with the quickly changing employment landscape and global ecosystem, it is becoming increasingly critical that children not only learn, but more importantly learn how to learn. Education thus, must move towards less content, and more towards learning about how to think critically and solve problems, how to be creative and multidisciplinary, and how to innovate, adapt, and absorb new material in novel and changing fields. Pedagogy must evolve to make education more experiential, holistic, integrated, inquiry-driven, discovery-oriented, learner-centred, discussion-based, flexible, and, of course, enjoyable."
Teaching science to young children is a vitally important role. Taking natural curiosity and engaging with it in a way that encourages learning requires not just dedication, but a good understanding of education theory and its application. If we are to have a prosperous, equitable future, we need good teachers who can impart scientific knowledge, focus intellect and nurture skills such as research, enquiry and problem-solving. These skills will be in high demand in the new economy and are central to us tackling challenges already identified and those yet to come. Whether it is our climate, our health, our ageing population, our food supply, our economy or our security, scientific discovery and the use of scientific knowledge will be at the core of our ability to respond. That is not to say that every child who learns science will go on to be a scientist and nor do we want them to be. But we need all children to develop more than a passing knowledge of how science works, of statistics and probabilities, and of the need to seek out the evidence behind assertions. An understanding of the history of science and the importance of the scientific method will allow children to grow into people who have important contributions to make to society. As the world continues to change at a rapid pace, science teaching must remain dynamic and reflect the latest and best techniques for guiding children’s exploration of that world. This book is an important resource for those who have been given the responsibility of teaching science. I wish the authors every success.
There is, and this is good for the country,
This is tied to human capital. It is something that I, as an economist, is particualry interetsed in this question. India's growth is going to be driven by human capital. The human capital plays out in innovation and drives growth and that it also adds to the creative potential of the individual to be able to understand the complex problems and then also solve the complex problems that economies are going to face. I also have a podcast on this with Professor Claudia Williamson where this has been discussed in greater depths.
I DID NOT HAVE ANY BACKGROUND IN MATHEMATICS AND STATISTICS. HOW WILL I BE ABLE TO COPE UP?
Having exposue to mathematics and statistics is neither a necessary condition nor a sufficient condition to be able to master the application of complete data sciences tookit. Of course, you need to know linear algebra and calculus to be able to acquire the mathematical understanding of the exercise of model building which is a step by step process. All the models that used for predictive analytics are based on some estimators - OLS estimtors being the most common, of couse. In the OLS estimators, the idea is to optimse the sum of squared residuals which, therefore, provides the best estimates of the explanatory variables.. The first derivative and second derivative is used in order to underscore the maximum values and the minimum values. Until and unless, the mathematics of the linear regression is not understood you are not going to be able to get a firm understanding of the non-linear regression A very important constituent of mathematics is that learning mathematics has shown to linked to enhance critical thinking and I have a separate blog on this. I would urge you to read the blog If that is the case then you might have to put in additional efforts. You will be able to cope up with it provided you are willing to work diligently with me. I will, before the course modules for you, who do not have the required mathematical knowledge take a mathematic referesher course lasting 2 hours and, in those 2 years, will. Mathematics is only a language and with any language you just need to practice it to be able to both understand it and then use it to speak what is in your mind. I would. The logic of most of the models are mathematicsl and those need to understood completely in order to understand its applicability
IS IT SUITABLE FOR ME GIVEN THAT I AM NOT A SOCIAL SCIENCE STUDENT. I MIGHT BE PURSUING LAW OR POLITICAL SCIENCE. WHAT SHOULD I DO?
Empirical reseach is putting the data into testing evidences. Fields such as law, sociology, Data is now being genrated in incresing quantity and with lot of ease. In case you are a law student or pursuing any other social science then the recommendation for you will be definately take the course on linear regression and time series analysis at the very least. Publication in top journals, in so far as I have seen, require the application of linear regression analysis and the application of linear regression is vast in that it also involves the application, dependiing upon the objective of the study, of dummy variables or interaction terms. The dropbox folder of linear These skills and the knowledge that any student would accumulate will apply both in academia and industry. In academia, obviously, I would argue that it is as importnat for you as it is importna for anyone who would be
I would make a strong case, as I have done in this blog, that the course on linear regression should be done by all. Most of the reseach questions in industry and academia are
data is being generated at a break neck
so, it is highly important that you have knowledge of the sophistictaed tool kit that
I would, most certianly, say that if you pursuing fields such as sociology, political science,
I DO NOT HAVE DOMAIN KNOWLEDGE OR BUSINESS KNOWLEDGE. SO, WHAT SHOULD I DO AT THE MOMENT?
I have designed practical taks for every module which, in hindight, do involve the application of business domain. Therefore, in the tasks that I give you, I would make sure that some elemente of business domain knowledge is covered in
I will give you some reaadings on the domain knowledge or the business knowledge which. At this stage, I would tend to make the argument that domain knowledge or business knowledge is something that you will acquire once you will enter the job market. The employers do not expect you to have full knowledge or understanding of the
One underlying message that comes from my podcast
I DO NOT PROGRAMMING OR CODING. HOW WILL I ABLE TO COPE UP WITH IT?
From the F&Q 1, it should by now be clear to you that software packages are one of the three parts that make up data sciences. As I have mentioned, there are a number of software packages that are used in order to run the models. Some of those softwares are STATA, E-Views, GRETL, JMulti and others too. I have a blog on this topic and it is titled review of different software packages. The blog gives some explanation of the pros and cons of the different software packages and so anyone who is curious, and you should be curious, then you should go and read this blog. As an aspiring data scientist, you should also be aware of all the softwares out there since each software has its own speciality that stands itself out. For instance, STATA, as such, is widely used for cross-sectional models and E-Views, as such, is widely used for the time series models. The only thing that you would need is to practice running the linear and the non-linear models R, Pyhton, and others require coding but quite a many softwares do not require coding. Now, somewhere
The most powerful asset you have is your ability to think and then implement that thinking in a model. Therefore, In all of these, I would most definately make a case that do not get caught thinking too much on the software packages. Bulk of your thinking should go into understanding the fundamentals of the models and this is what you will be paid for in your work as a data scientist. There are professional programmers and coders out there who do not have much undertanding of the model building which. Since the field, as such, is exploding so you are not going to be an expert in everything so
WHAT IS ECONOMTERICS HAS TO DO WITH DATA SCIENCES?
It has everythng to do with data science. The recent winning of the Nobel Prize in Economic Sciences the year 2021 just underpins this and much of the contribution for which the economists won the Nobel Prize for was for understanding the causality and concepts such as these. An economterician, naturally, is a data scientist and this is precisely the reason many corporations have begun to
´Where the two do differ though, is the outcome and the usage of the models they create. An econometrician and a data scientist could produce the exact same model and write up very different discussion papers.
´I’ve noticed that most data scientists seem happy with their model solely if it gets a high accuracy score. We all know the issue with this — especially in models trying to predict rare events. Econometrics tends to look deeper into the model and attempt to deduce causality.
´While data science is focused solely on prediction, econometrics attempts to achieve high accuracy while also seeking to find causal relationships. finding causal relationships is a constraint, however, one that will often harm your accuracy.
´Econometrics trades off accuracy for understanding.
´Data science uses data to find relationships, Econometrics uses data to prove relationships. Econometrics starts with a theory and then uses a model to test its assumptions.
One big advantage of knowing economterics is that it is grounded in theory. Therefore, since it is grounded in theory, it therefore forces you think about the relationship that might exist between socio-economic forces.
If you are very curious, watch this video.
WHAT JOB OPPORTUNITIES CAN I HAVE IF I DO THE COURSES ON ELEMENTARY COURSES?
Opportunities would show themselves off automatically. At this stage, I would say that you should only focus on learninb both the theoritical background and the way the things are praticall implemented. In order to be able to achieve this, I have, from the inputs of the practitioners, constructed industry that closely mirrors and replicates the industry cases. I would give you industry cases to be able to work on them
Therefore, get the maximum exposure from these and once you are trained, you will
HOW DIFFERENT ARE THE ADVANCED TECHNICAL COURSES FROM THE COURSE ON LINEAR REGRESSION, TIME SERIES ANALYSIS AND THE NON-LINEAR REGRESSION?
As I have argued that the foundation of data sciences is the course on linear regression, time series regression, non-linear regression and cross-sectional models. Even in here, the course on non-linear regression and cross-sectional models are relatively more advanced than the course on linear regression and time series. I have a blog post on why the course on linear regression is for everyone.
Given this, coming back to the other advanced courses, these courses are highly advanced. They are very different and highly advanced. Having said this, I must point out that the course of applied linear regression is the foundation course. The applied
Most complex ideas, both in reserch and industry, are evalued with the use of the linear regression models. So, it is very importnat to be able to
I AM AN ENGINEERING STUDENT. SHOULD I THINK ABOUT ENROLLING FOR THE COURSES.
Most engineering department of courses do not cover the basic of the statistics and probability.
One advantage that an engineer has is the ability to code and program. I would argue that this does not mean that any enginner has a upper hand. You only have an upper hand only if you know the fundamentals of mathemtics and statistics. Economterics asa Though it is an importnat skill
Follow us:
Contact Details
© 2023 All rights reserved. Academy for Data Sciences.