Ryerson MBA is participating in The Economist’s Real Vision Investment Case Study against teams from some of the world’s other leading MBA programs. Their challenge? Analyze Walmart and Amazon to determine which retail giant’s stock is the better investment choice over the next 10 years.
Here, team member Saad Rahman discusses how the team used models to assist in their analysis. Visit The Economist website to view the full analysis and vote for Ryerson MBA.
What are models?
Strictly speaking mathematical models are “abstract models that use mathematical language to describe the behaviour of a system.” As social professor Scott Page would put it, “You model so that you can be an intelligent citizen of the world.” Models help us develop a clearer understanding of the world around us and help us also make more informed decisions. Though there can be many factors to consider and the models may not be the most accurate, they tie us down a little bit and help us achieve a more realistic understanding of the topic at hand.
Why did we use models in our analysis?
So why was the model so relevant in what we were doing? Well for one thing, there was a lot of qualitative data that need to be quantified, and even the quantitative data was reflecting a certain pattern which was captured through various other analysis. To account for these two movements, we had to “tie down to the mast.” The biggest challenge for us was that the prediction needed to be made 10 years out. Normally when using statistical data, a maximum of 2–3 years projections can be made. As the time line increases so does the uncertainty associated with them. So the major issue was how to incorporate the long-term trend and possible return. So early on in the project, we realized that we would need some kind of model, but just incorporating quantitative data would not be enough. So we came up a somewhat unique way to also incorporate the qualitative data. This qualitative data was basically based around the trends in the industry and how they were likely to evolve in the future.
How did we use the models?
The process combined two models: a regression analysis model compared company share prices to economic indicators, generating coefficients for a share price calculator. Model 2 used a three-year Discounted Cash Flow analysis and a Weighted Comparative Impact Model to factor in anticipated performance based on retail trends for the subsequent seven years. Both predictions were combined to forecast share prices.
So essentially there were three sub-models:
- Quantitative Analysis
- Qualitative Analysis
- What were the industry trends
- How the companies were responding to those trends
But then we had to give relevant weightages, so we created two models and in the end combined their results in a certain weightage.
To accomplish this, a qualitative analysis of industry and academic literature was performed to evaluate the inherent range of potential future outcomes for three segments: macroeconomic environment, retail sector trends, and each company’s business strategy in relation to those trends.
Each of these segments was broken down and quantified, producing two different models (Model 1 and Model 2) upon which we based a price prediction for each company in 2025.
Want to read more about our models and how they influenced our analysis? Visit The Economist’s Real Vision Investment Case Study to view our full analysis. And don’t forget to vote for us for the People’s Choice Award – registered users can vote daily!