Forecasting can be found in every corner of the business world today. When done in tandem with accurate time series analysis, it enables sound prediction of future values. In this course, you will explore the use of time series analysis and the four components of time series data. Consider, there are a number of time series that may require forecasting but do not have any discernible trend, such as a stable product environment or a very short timeframe. In this course you will continue exploring forecasting by examining stationary time series and the situations in which they most often occur and practice forecasting techniques and stationary time series analysis. You will then examine stationary data where no substantial change is taking place. Lastly, you will move to data that is changing. A layer of complexity can be added to forecasting in the form of seasonality, where the time series being studied regularly changes with each season. This added element must be considered in any prediction of future periods.
It is recommended to only take this course if you have completed Presenting Quantitative Data, Descriptive Statistics for Business, Making Predictions Using Statistical Probability, Inferential Statistics, and Multivariable Comparisons or have equivalent experience.
KEY COURSE TAKEAWAYS
Explore the use of time series analysis and the four components of time series data
Discuss the role of forecasting in your organization
Apply the tools of analysis, including graphs and diagnostic measures
Explore forecasting by examining stationary time series and the situations in which they most often occur
Define autocorrelation and the key role that it plays in forecasting stationary data
Practice forecasting techniques and stationary time series analysis
Analyze linear data: data that either increases or decreases at a steady rate
Observe forecasting performed with an original data source and a single variable
Explore the relationship between two time series and the methods used for these projections
Consider curvilinear models and then develop your own time series forecasts
Define the factors that cause seasonality
Identify the steps to follow in forecasting a seasonal time series, which include quantifying and separating seasonal influences and then applying that information to your forecasting
Follow those steps yourself in computing seasonal forecasts
Cindy van Es is professor of practice in the Dyson School of Applied Economics and Management. She has a PhD in statistics from Iowa State University, and joined Cornell in 1988. She teaches three courses in the undergraduate business program: Introductory Statistics, Business Statistics, and Impact Learning: South Africa. Her general area of interest is statistical education, with a focus on business applications and teaching through social justice examples.
She currently serves as director of Dyson’s Undergraduate Business Program. In this position, she provides strategic leadership and supervision on activities within the undergraduate program at the school, focusing specifically on implementation of the undergraduate curriculum and review of academic policies