Time Series in R - Part 1
Estimated Time Commitment: 3 Hours
In this course we will introduce analytical methods to analyze time series data to build forecasting models and support decision-making. You will first learn the importance of moving averages for removing white noise and identifying trends. Building on concepts of moving average, you will be introduced to exponential smoothing techniques that place greater weight on more recent time series observations.
While the moving average and simple exponential smoothing techniques are helpful for removing white noise, they often are not helpful for creating forecasts because they do not account for long-term trends and seasonal fluctuations in time series. For this reason, you will explore Holt’s exponential smoothing and Holt-Winters forecasting models to create more accurate forecasts.
Finally, you will be introduced to autoregressive models, which are similar to “regular” linear regression models except that the explanatory variables are observations from prior periods, rather than other variables.
Upon successful completion, you will be able to:
Examine moving averages and exponential smoothing in relation to time-series data
Identify three types of exponential smoothing methods: simple, Holt, and Holt-Winters
Execute the three types of exponential smoothing methods for creating forecasts of time-series data
Develop an understanding about autoregression, knowing its role in analyzing time- series data
Introduction Video: Jose Rodriguez - Forecasting Models in Practice
Introduction to the Skill
Glossary
Moving Average
Moving Averages -- R Example
Knowledge Check 1
Simple Exponential Smoothing
Simple Exponential Smoothing -- R Example
Knowledge Check 2
Holt's Exponential Smoothing
Holt-Winters Forecasting Model
Holt-Winters Forecasting Model -- R Example
Knowledge Check 3
Autoregression
Autoregression -- R Example
Knowledge Check 4
Instructions
Exercise Files
Debriefing
Conclusion
Final Quiz
Survey Instructions
Feedback Survey
Survey Verification
Next Steps