Regression Analysis – An understanding and application
“There are two ways of lying. One, not telling the truth and the other, making up statistics – Josephina Vazquez Mota”
Statistics is a very decisive part of data analysis. While there is so much talk about the right data, it is equally important to have the proper analysis to assess the results. As much as we speak about data driven decisions, we should also bear in mind that deconstructing the data is equally vital. Hence interpretation and analysis play a very crucial role in data analysis.
Regression analysis is a simple method for investigating functional relationship between variables. It is a way of sorting out the degree of impact of a variable. It answers questions relating to which variables are important, which are the variables that can be ignored and how they interact with one another etc. It is a statistical tool used to understand and quantify the relation between two variables. Regression can also be used to test the strength of relationship between variables and also help model the future relationships that may arise.
In business, regression is used mainly for explaining a phenomenon, predicting the future or to decide what is the next course of action. Primarily regression is used in business for forecasting and optimization
1. Predictive analysis or Predicting the future – The most common usage of regression is forecasting the uncertain future. Here future opportunities and risks are forecast to a greater degree of accuracy
2. Optimization of business – Regression enhances the productivity of a business and also helps in rapid economic advancements. This kind of data driven analysis helps to avoid guess work and helps make well-crafted business decisions.
3. Supports decisions – Regression helps adopt a scientific approach to decision making. It helps us analyze data that are significant to our decisions.
4. Spot errors – Regression helps to spot and correct errors as they help analyze and make rational decisions
5. New Insights – They help gather data to give greater insights to unexplored or unseen avenues
Types of regression
1. Linear – Simple linear regression model assesses the relationship between a dependent and an independent variable
2. Multiple Linear – It is similar to linear regression model with the exception that multiple independent variables are used in this model
3. Non-Linear – They are used for more complicated data sets in which dependent and independent variables show a non-linear relationship
Regression is used to analyze the marketing effectiveness, pricing and promotion on sales of a product etc. It can capture isolated or combined impact of an event. It can be used to assess the risk in financial services. It can be used to calculate the BETA (volatility of returns relative to the overall market) for a stock.
They have been widely used in the following areas:
1. Agricultural sciences
2. Industrial and Labor relations
5. Environmental science
6. Industrial production
7. Healthcare etc.
Regression analysis is used to uncover patterns and relationship that were previously undiscovered and unnoticed. They help not only in making good management decisions but also help uncover lapses in judgement.
“Most people use statistics like a drunk man uses a lamp post, more for support than for illumination – Andrew Lang”
Author – Benila Jacob, Mark John