In this video, the instructor helps you find questions to be answered with data.
- [Instructor] Business intelligence is useful for making a wide variety of forecasts, and helping businesses to be more efficient in running their operations. But not all questions can be answered by data realistically. Some questions are inherently subjective and don't require data to come up with an answer. Democrat or republican? Red Sox or Yankees? What color is best? Where should we live? These are questions that are inherently based on subjective opinions and there's no data that we can gather that could help us to make a decision.
But even when we have questions that could be answered objectively in theory, sometimes we don't have data available or we don't have the right kind of data. We refer to this as omitted variables bias. In particular, if we think about questions on forecasting sales for example, if we're in a totally new environment and we've never seen any environment like this before, it's very difficult to gather data. If our company has expanded internationally for the first time selling a completely new product, we won't have any data available that can help us with making forecasts on sales.
There's several considerations we need to take into account, in fact, when picking a question to answer with a forecast. First is this a material issue for the company? In particular, making forecasts with data is difficult. It takes time, takes resources, and it takes effort. If answering the question isn't worth dedicating those resources, then we should move on. Not all forecasts are worth doing. Second, we need to ask ourselves, is this a nonsubjective question? Democrat or republican, Yankees or Red Sox, these are obvious examples of subjective questions.
But in some cases, we have examples of questions that aren't so subjective. For example, which product should we consider expanding to next? Well, we can gather data based on that, but there's an element of subjectivity. What market do we feel like is going to be most appropriate to our customers? There's an element of subjectivity here that might lead us to be unable to make an appropriate forecast. Third, we need to ask ourselves is data available? If we're expanding to a new overseas market offering a new product, there might not be any data available.
Or we might have limitations on the availability of data, simply because it's never been collected before. There's a lot of potential data out there that could be collected that is not today. And finally, we need to ask ourselves is it feasible to alter the status quo? If we're talking about making a forecast on the optimal pricing for a company for example, but our firm is not able technically to offer different prices to different customers, we simply lack the technology to do that, well then we can make all the forecasts we want, but it's not likely to be useful to the firm at the end of the day.
Join Professor Michael McDonald and discover how to use predictive analytics to forecast key performance indicators of interest, such as quarterly sales, projected cash flow, or even optimized product pricing. All you need is Microsoft Excel. Michael uses the built-in formulas, functions, and calculations to perform regression analysis, calculate confidence intervals, and stress test your results. You'll walk away from the course able to immediately begin creating forecasts for your own business needs.
- Understanding big data and predictive analytics
- Gathering financial data
- Cleaning up your data
- Calculating key financial metrics
- Using regression analysis for business-specific forecasts
- Performing scenario analysis
- Calculating confidence intervals
- Stress testing