Multivariate linear regression using statsmodels and sklearn

I do have to admit that statsmodels is one Python library that I have not spent a lot of time exploring. I am much more familiar with sklearn, which is the reason why most of my posts involve employing this library in some fashion. Therefore, when I took a Coursera course, Python and Statistics for Financial Analysis, I jumped at the chance to learn how to use Python to carry out financial analysis. There was a portion of the course that covered linear regression, which I have a keen interest in. The last week of the course discussed simple univariate linear regression and the final part of the course covered multivariate linear regression. My most recent post concerning linear regression can be found here:- https://medium.com/mlearning-ai/a-case-of-simple-linear-regression-using-sklearn-statsmodels-d2160d456994

In this post I will endeavour to discuss multivariable linear regression using statsmodels and sklearn. Multivariate regression is a regression model that estimates a single regression model with more than one outcome variable. It goes without saying that multivariate linear regression is more difficult to estimate than univariate linear regression because there are more input variables to consider.

I have coded a program to make predictions on a multivariate dataset that I created using sklearn’s make regression function, and intend to discuss the script.

I have written the program in Google Colab, which is part of the Google suite of free online applications. Google Colab is a great Jupyter Notebook to use because the fact that the program is used online means that it is portable and can be used on any computer that has access to the internet and Google. The only main drawback that I have seen working with this program is the fact that it does not have an undo function, so care must be taken not to overwrite or delete valuable code.

When I created the program, I imported the libraries I would need to execute it. I generally only import libraries as I need them, but in this instance I have imported pandas, numpy, sklearn, statsmodels, matplotlib and seaborn. Pandas is used to create and manipulate dataframes, numpy creates arrays and performs algebraic computations, sklearn houses the varied functions used in machine learning…

Related posts:

Em fevereiro de 2018, a Sweek Brasil lançou um desafio de microcontos chamado #microcoração. Apesar do dia dos namorados ser celebrado apenas em 12 de junho (em nosso país), a data é comemorada no…

It is crucial to ensure that any action taken to address climate change which is a transboundary crisis, is inclusive and equitable, benefiting all communities, particularly those in the Global South…

How does my family history connect to my American creed you might ask? My mother efforts to be in the United States connects to my American creed. My mother worked hard to come to the United States…