Data science toolkit is vast. To get a sense of how vast, click here. As I have argued in my podcasts and also in my previous, data sciences and machine learning is vast and the toolkits that are being used are constantly expanding. Any novice is just likely to get lost and be dumbfounded and might even think where should I start if I want to make a career in data science. I would argue that the fundamental building block for all of machine learning, data science and econometric toolkit is the course on linear regression. There are quite a few important reasons as to why this is the case. In this blog, I will make an attempt in putting forth a persuasive argument in order to defend my thesis.
There are five arguments that I would like to present here in order to drive home the point that I am trying to make here.
From the perusal of the data science book of class 12, I see that it is application of
The new education policy
There are a number of arguments that I would like to present here that would tend to point to a case that the course of linear regression should be done by all. There are a lot of tangi
Therefore, getting your fundaments strong in the use and application of linear regression case is extremely important. For students, who want to make a career in the big data and
Essentially, in a linear regression, the idea is that you mode the relationship between
This is something that I have also understood. The domain knowledge takes years to acquire and that too if you are constantly reading about it.
Even in industry, the
1. For all the complex analysis, in the time series, the starting building block of all the models is the linear regression analysis. All the complex ideas, in social sciences and even in industry, are tested with the use of this model. The software that will be used here, that is GRETL. To be able to understand, the use of dynamic models – which are nothing but the time series models – it is quite fundamental that one is able to understand the application of linear regression models. When I say, the fundamental, I mean even the three diagnostic tests that are multicollinearity, heteroskedasticity and autocorrelation. For instance, to be able to understand the application of time series models, it is imperative that one understand the linear regression models and what are its shortcomings. To be able to understand the concept of non-stationarity – which is also called unit root – it is important to be
2. It must be clear to you that the linear There are a number of estimators in econometrics and data science. They are, for example, to name some, the two stage least square, maximum-likelihood estimators. The starting foundation of all them is the Ordinary Least Square Estimator.
3. I would tend to argue that most complex ideas in the area of social sciences and, even in industry, can be implemented with the help of the linear regression. Most journal articles published in the top tier journals used this methodology. Therefore, it is of utmost importance that one is able to understand the application of this technique. Perhaps, the most important application of the linear regression is in the area of pricing. Pricing the product is one of the most important challenge facing any organization and company and now with so much data being collected, it is but imperative that
This is an excellent blog on the technicalities involved when the linear regression model is used in the analysis of pricing.
4. The linear regression is also able to capture the non-linearities once the dependent variable is modified
By transforming the explanatory variable into something that does have a linear relationship with the outcome and entering that transformed variable into our model we can maintain the assumption of linearity.
As I write in M.Phil thesis that I wrote that the
As India moves in to getting more digital, it is but natural that more and more data is going to be generated
The basis of understanding the machine learning algorithm is the linear regression. The course on linear regression is the bedrock of any
As an economist, I can argue that t
I would make the
If you have the skills, you can sit in India and export your skills to firms and companies that require the use of this skill
Comments