Skip to main content

Forecasting in R

Learn how to make predictions about the future using time series forecasting in R including ARIMA models and exponential smoothing methods.

Start Course for Free
5 Hours18 Videos55 Exercises45,720 Learners

Create Your Free Account

GoogleLinkedInFacebook

or

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.

Loved by learners at thousands of companies

Course Description

Use Forecasting in R for Data-Driven Decision Making

This course provides an introduction to time series forecasting using R.

Forecasting involves making predictions about the future. It is required in many situations, such as deciding whether to build another power generation plant in the next ten years or scheduling staff in a call center next week.

Forecasts may be needed several years in advance (for the case of capital investments), or only a few minutes beforehand (for telecommunication routing). Whatever the circumstances or time horizons involved, reliable forecasting is essential to good data-driven decision-making.

Build Accurate Forecast Models with ARIMA and Exponential Smoothing

You’ll start this course by creating time series objects in R to plot your data and discover trends, seasonality, and repeated cycles. You’ll be introduced to the concept of white noise and look at how you can conduct a Ljung-Box test to confirm randomness before moving on to the next chapter, which details benchmarking methods and forecast accuracy.

Being able to test and measure your forecast accuracy is essential for developing usable models. This course reviews a variety of methods before diving into exponential smoothing and ARIMA models, which are two of the most widely-used approaches to time series forecasting.

Before you complete the course, you’ll learn how to use advanced ARIMA models to include additional information in them, such as holidays and competitor activity.
  1. 1

    Exploring and visualizing time series in R

    Free

    The first thing to do in any data analysis task is to plot the data. Graphs enable many features of the data to be visualized, including patterns, unusual observations, and changes over time. The features that are seen in plots of the data must then be incorporated, as far as possible, into the forecasting methods to be used.

    Play Chapter Now
    Welcome to Forecasting Using R
    50 xp
    Creating time series objects in R
    100 xp
    Time series plots
    100 xp
    Seasonal plots
    100 xp
    Trends, seasonality, and cyclicity
    50 xp
    Autocorrelation of non-seasonal time series
    100 xp
    Autocorrelation of seasonal and cyclic time series
    100 xp
    Match the ACF to the time series
    50 xp
    White noise
    50 xp
    Stock prices and white noise
    100 xp
  2. 2

    Benchmark methods and forecast accuracy

    In this chapter, you will learn general tools that are useful for many different forecasting situations. It will describe some methods for benchmark forecasting, methods for checking whether a forecasting method has adequately utilized the available information, and methods for measuring forecast accuracy. Each of the tools discussed in this chapter will be used repeatedly in subsequent chapters as you develop and explore a range of forecasting methods.

    Play Chapter Now
    Forecasts and potential futures
    50 xp
    Naive forecasting methods
    100 xp
    Fitted values and residuals
    50 xp
    Checking time series residuals
    100 xp
    Training and test sets
    50 xp
    Evaluating forecast accuracy of non-seasonal methods
    100 xp
    Evaluating forecast accuracy of seasonal methods
    100 xp
    Do I have a good forecasting model?
    50 xp
    Time series cross-validation
    50 xp
    Using tsCV() for time series cross-validation
    100 xp
  3. 3

    Exponential smoothing

    Forecasts produced using exponential smoothing methods are weighted averages of past observations, with the weights decaying exponentially as the observations get older. In other words, the more recent the observation, the higher the associated weight. This framework generates reliable forecasts quickly and for a wide range of time series, which is a great advantage and of major importance to applications in business.

    Play Chapter Now
    Exponentially weighted forecasts
    50 xp
    Simple exponential smoothing
    100 xp
    SES vs naive
    100 xp
    Exponential smoothing methods with trend
    50 xp
    Holt's trend methods
    100 xp
    Exponential smoothing methods with trend and seasonality
    50 xp
    Holt-Winters with monthly data
    100 xp
    Holt-Winters method with daily data
    100 xp
    State space models for exponential smoothing
    50 xp
    Automatic forecasting with exponential smoothing
    100 xp
    ETS vs seasonal naive
    100 xp
    Match the models to the time series
    50 xp
    When does ETS fail?
    100 xp
  4. 4

    Forecasting with ARIMA models

    ARIMA models provide another approach to time series forecasting. Exponential smoothing and ARIMA models are the two most widely-used approaches to time series forecasting, and provide complementary approaches to the problem. While exponential smoothing models are based on a description of the trend and seasonality in the data, ARIMA models aim to describe the autocorrelations in the data.

    Play Chapter Now
    Transformations for variance stabilization
    50 xp
    Box-Cox transformations for time series
    100 xp
    Non-seasonal differencing for stationarity
    100 xp
    Seasonal differencing for stationarity
    100 xp
    ARIMA models
    50 xp
    Automatic ARIMA models for non-seasonal time series
    100 xp
    Forecasting with ARIMA models
    100 xp
    Comparing auto.arima() and ets() on non-seasonal data
    100 xp
    Seasonal ARIMA models
    50 xp
    Automatic ARIMA models for seasonal time series
    100 xp
    Exploring auto.arima() options
    100 xp
    Comparing auto.arima() and ets() on seasonal data
    100 xp
  5. 5

    Advanced methods

    The time series models in the previous chapters work well for many time series, but they are often not good for weekly or hourly data, and they do not allow for the inclusion of other information such as the effects of holidays, competitor activity, changes in the law, etc. In this chapter, you will look at some methods that handle more complicated seasonality, and you consider how to extend ARIMA models in order to allow other information to be included in the them.

    Play Chapter Now
    Dynamic regression
    50 xp
    Forecasting sales allowing for advertising expenditure
    100 xp
    Forecasting electricity demand
    100 xp
    Dynamic harmonic regression
    50 xp
    Forecasting weekly data
    100 xp
    Harmonic regression for multiple seasonality
    100 xp
    Forecasting call bookings
    100 xp
    TBATS models
    50 xp
    TBATS models for electricity demand
    100 xp
    Your future in forecasting!
    50 xp

In the following tracks

Quantitative Analyst with RTime Series with R

Datasets

Excelfile in the first exercise

Collaborators

Lore Dirick
Davis Vaughan

Prerequisites

Time Series Analysis in R
Rob J. Hyndman HeadshotRob J. Hyndman

Professor of Statistics at Monash University

Rob J. Hyndman is Professor of Statistics at Monash University, Australia, and Editor-in-Chief of the International Journal of Forecasting. Rob is the author of over 150 research papers and books in statistical science. In 2007, he received the Moran medal from the Australian Academy of Science for his contributions to statistical research, especially in the area of statistical forecasting. He is the author of about 20 R packages, including the popular forecast package.
See More

What do other learners have to say?

    FAQs

    Join over 11 million learners and start Forecasting in R today!

    Create Your Free Account

    GoogleLinkedInFacebook

    or

    By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.