In this video, the instructor creates a forecast based on a regression analysis.
- [Narrator] Once we've gone through, figured out what our question is, gathered our data, cleaned it up, and ran our analysis, then we're ready to test our options. Now, the reality is that big data today is generally used for monitoring and identifying problems. And the reality is that that's great, but it's really not fully capturing the power that data has. It's not using data to make forecasts the way we've talked about so far. In fact, in many business cases today, there's limited use of data for making decisions thus far.
Best practice at most firms is to allow tests on variables based on our decisions. In particular, we want to gather that data and then figure out how to harness it to make choices. For example, one classic business question that almost every firm faces is how much marketing should we use on our product? What is the optimal amount of marketing spend? In particular, we might ask, how is profit going to be impacted if our marketing spend is doubled? This is exactly the kind of question that data can help us with.
So when it comes to making decisions based on data, data can be the basis for objective choices whether it's making choices like how much marketing spend should we use or forecasting investment returns on a particular project or pricing optimizations or resource allocations. All of these types of questions are best going to be answered with data. Once we have gathered the appropriate data, we need to test our choices based on the data that we have.
Data forecasting doesn't have to be hard. As we've seen already, we can start with Excel and we can run some kind of a regression analysis. For example, we could regress sales on our other variables or we could regress cash holdings at the end of the day on the employees who are working. Whatever regression we run, we need to do a hedonic forecast after making assumptions about other variables in the future. What do I mean by hedonic forecast? Well, a hedonic forecast takes the coefficients from our regression and puts in assumptions along with those coefficients to spit out an outcome that represents our prediction.
We'll take a look at how to use this in Excel coming up.
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