Target group
Analysts with a quantitative background as well as non-statistical analysts and domain experts who would like to augment their time series modeling proficiency
Course structure
Introduction to Time Series
- Defining a time series.
- Using the TIMESERIES procedure to transform transactional data into time series data.
- Defining and exploring the systematic components in a time series.
- Describing the decomposition of time series variation.
- Listing three families of time series models.
- Introducing SAS Studio.
- Introducing the concepts of white noise and autocorrelation.
Exponential Smoothing Models
- Exploring weighted average models and exponential smoothing.
- Comparing and contrasting simple mean, random walk, and exponential smoothing models.
- Imputing missing values within a time series.
ARIMAX Models
- Differentiating between ARMA and ARIMA models.
- Defining a stationary time series and identifying its importance.
- Describing and identifying autoregressive and moving average processes.
- Defining the differences between a random walk series, a white noise series, and an autoregressive (AR) series.
- Estimating autoregressive parameters .
- ARMAX and time series regression.
- Accuracy and forecasting of ARIMAX.
Unobserved Components Models
- Introducing unobserved components models (UCM) and focus on the multiple sources of error and parameters as a function of time.
- Describing the basic component models: level, slope, seasonal.
- Exploring the UCM model parameters.
- Running a UCM model using the UCM procedure.
- Defining Random Walk and Linear Trend series.
- Building a UCM model.
