Regression Analysis For Business Insight - Consulting Case

Learn how to use regression analysis to answer business questions using a consulting engagement scenario

What you'll learn

Learn how to answer real world business questions using model results

Learn how to conduct residual analysis to test for autocorrelation, heteroscedasticity and normality and validate model

Brief theoretical overview of OLS and the Gauss Markov Theorem

Learn how data is collected in a typical consulting engagement

Understand the difference between causality and correlation

Learn how to conduct seasonality analysis for your dependent variable

Understand model statistics (R-SQ, Durbin Watson and MAPE) and how that is used to evaluate the model

Learn how to decomp the model to get the daily incremental values for each explanatory variable

Understand how to use the Actual vs Predicted graph to improve the model

Learn how to effectively use essential excel formulas (Index Match, Sumifs, RSQ) and build graphs (Line Graphs, Scatter Plots, Histogram)

Requirements

No requirements or prerequisites. All course content such as workbook, syntax files and raw data file will be provided

Description

In this course, we will be combining the practical and theoretical aspects of regression analysis by building a ordinary least squares model to answer real world business questions then validating it using the Gauss Markov Theorem to ensure it is theoretically accurate as well.We will be taking the place of a management consultant who has been tasked by the client to find the key drivers of their sales using a regression model and then answer business questions has using that model.These are a few of the questions the client has asked us to answer:What was the ROI from marketing efforts in 2022? What was the impact of competitor marketing in 2022? What was the impact of the increase in price of our product?What is the price elasticity of demand for our product? Which month had the highest baseline sales?We expect our marketing to be 10% more efficient next year. If we keep spend at the same level, how many incremental units will we get from our marketing efforts next year? We would like to increase price by 10% next year, what impact will that have on our business? In addition to this, we will also briefly go over the theory of ordinary least squares regression and the Gauss Markov theorem. Which we will then put into practice to validate our model by analyzing the residuals. We will also learn how to evaluate our model using key statistics like the R Square, Durbin Watson and the MAPE.All analysis will be done in excel, so we will learn how to setup a model workbook in excel and learn how to use essential functions such as Index Match, Sumifs and Countifs. We will also go over how to build line graphs, scatterplots and histograms.This course is ideal for anyone interested in working in consulting, data analytics, data science, econometrics and academic research.

Overview

Section 1: Intro: What does this course cover?

Lecture 1 Introduction: What does this course cover

Section 2: OLS and Gauss Markov Theorem Overview

Lecture 2 OLS and Gauss Markov Overview

Section 3: Data exploration and Seasonality analysis

Lecture 3 Project Intro and Business Questions

Lecture 4 Data Definitions

Lecture 5 Causality vs Correlation

Lecture 6 Seasonality Analysis

Section 4: Model Decomp and Analysis

Lecture 7 Regression Model Formula

Lecture 8 Model Output and Causality Check

Lecture 9 Model Decomp

Lecture 10 Actual vs Predicted Graph

Section 5: Residual Analysis

Lecture 11 Residual Analysis

Lecture 12 Residual Analysis For Preview

Section 6: Answering our clients questions

Lecture 13 Business Questions Part 1

Lecture 14 Business Questions Part 2: Marketing Questions

Lecture 15 Business Questions Part 3: Pricing Questions

Anyone interested in learning about regression analysis,Consulting,Data Analytics/ Data Science,Economist & Econometricians,Academic Research