Demand Forecasting with R

Online week long course for demand planners and data scientists

Who is the course for?

Demand planners

Experts and novices will learn the principles of demand forecasting and how to put them into practice

Data scientists

You will learn how to develop models and forecast using R. Appropriate methods of data analysis and forecasting will be demonstrated.

Business analysts

You will learn which models to use in different circumstances and how to select the most appropriate

What makes this course so special?

We teach forecasting fundamentals

You will learn how to model real life problems, not just how to create R code

We show how to do that in R

You will understand why specific functions are used to address specific problems

We know practicalities of forecasting

This forecasting course is not taught in a vacuum; we know what problems you face in practice and we will go beyond R

Course content

The topics for the course will be selected from the list below, depending on the audience and their preferences:

Topic 1

Forecasting principles

  • What to do and what not to do in forecasting,
  • Evaluating forecasting accuracy via error measures,
  • Uncertainty, prediction intervals and their evaluation,
  • How to inform decisions based on forecasting.

Topic 2

Time series components

  • Classical time series decomposition,
  • Simple forecasting methods (Naïve, Global Average, Moving Average),
  • Exponential smoothing,
  • Introduction to the ETS model,
  • Holt, Holt-Winters and Damped trend methods and their connection with ETS.

Topic 3

Linear regression

  • Simple linear regression,
  • Multiple linear regression,
  • Regression diagnostics,
  • Transformation of variables,
  • Variables selection,
  • Using regression in promotional modelling.

Topic 4

Advanced modelling approaches

  • ETS with explanatory variables,
  • Multiple frequencies,
  • Model and forecast selection,
  • Combination of forecasts,
  • Judgment and organisational aspects of forecasting.

Topic 5

Advanced forecasting methods and models

  • Intermittent demand forecasting,
  • ARIMA,
  • Hierarchical forecasting: cross-sectional and temporal hierarchies.

Overall, the course will last for a work week, 3 hours per day.

Learning outcomes

By doing this course you will be able to:

  • Understand how forecasting models work;
  • Understand what parameters of models mean;
  • Analyse time series structure;
  • Identify time series components;
  • Produce point forecasts and prediction intervals for any time series;
  • Evaluate the accuracy of different forecasting methods;
  • Make relevant managerial decisions based on the point and interval forecasts.
Meet your tutors
Prof. Robert Fildes

Robert Fildes is the Director of Centre for Marketing Analytics and Forecasting, Lancaster University Management School. He is a co-founder of International Institute of Forecasters and International Journal for Forecasting. He is an expert in judgmental forecasting methods and has an outstanding experience in making forecasting work for you.

Prof. John Boylan

John is Professor of Business Analytics at Lancaster University. His research interests are focused on supply-chain forecasting and include intermittence, seasonality and information sharing. He is an Editor-in-Chief of the Journal of the Operational Research Society and President of the International Society for Inventory Research.

Prof. Nikos Kourentzes

Nikolaos Kourentzes is a professor at Skövde University in Sweden, and a CMAF veteran. His research is on temporal aggregation, hierarchical forecasting, model combination and selection, with statistical or artificial intelligence methods. He works closely with industry, with an aim to help practice adopt and use state-of-the-art analytics. He is a world leading expert in ice creams.

Dr. Ivan Svetunkov

Ivan is an expert in area of statistical methods for forecasting. His areas of interest include intermittent demand, retail, forecasting NHS-related processes (e.g. A&E attendance) and others. He is an author and maintainer of greybox and smooth packages for R, implementing regression, exponential smoothing and ARIMA for forecasting purposes.

Dr. Sven Crone

Sven is expert in Operational Research and Information Systems in the domains of Demand Planning within Inventory Management for Supply Chain Management & Operations Management. He is a world-known specialist in neural networks forecasting and in forecasting support systems development.

Testimonials
AK
*
Athanasios Kontinopoulos
Economic Analysis and Research Department, Bank of Greece
I found the course really helpful, as it not only covered both basic and advanced concepts in forecasting, but it was also full of examples in R. The tutors were a strong asset, as they were very knowledgeable and provided great feedback. Strongly re... <p>I found the course really helpful, as it not only covered both basic and advanced concepts in forecasting, but it was also full of examples in R. The tutors were a strong asset, as they were very knowledgeable and provided great feedback. Strongly recommended!</p>
Read more
EL
*
Eric Ling
Head Of Forecasting, Supply Division of Howdens Joinery Ltd
https://www.howdens.com/
Following on from an audit that the team from Lancaster University conducted for us on how we were using SAP APO Demand Planning, they developed a tailored workshop for our entire Forecasting Team on Statistical Forecast Modelling Techniques. As a di... <p>Following on from an audit that the team from Lancaster University conducted for us on how we were using SAP APO Demand Planning, they developed a tailored workshop for our entire Forecasting Team on Statistical Forecast Modelling Techniques. As a direct consequence of the workshop we re-configured the SAP APO default settings on our statistical forecast models and the Forecasting Team were able to confidently apply different forecast models to different types of products. This has enabled our team of forecasters to manage a much greater number of statistical forecast models than previously, as well as allow them to focus their time and efforts on the products which they can add most value to. I would have no hesitation in recommending the Lancaster University training material to anyone looking to gain a more thorough insight into statistical forecast modelling.</p>
Read more
The schedule of the online courses

We plan to hold several courses over the year. So far, we have planned the course for the following dates:

  • 27th September - 1st October, 2021
  • 25th - 29th April, 2022
There are three options for participation
Basic plan
Access to the slides of the course
Access to all workshop materials
Communication with course tutors during the course

£750

Group plan
All the benefits of the basic plan
Discount for 3 or more participants

£500 per person

Turbo plan
All benefits of the basic plan
Support for a short project immediately after the training
International Institute of Forecasters certificate awarded in case of successful completion of the course. Ask us for details about this.
Physical copy of textbook of Ord, Fildes & Kourentzes (2017)

£1000

Course prerequisites

The only requirement for the attendees of the course is to know the basics of R, understand the "if-else" logic and have a willingness to do perform basic programming tasks. The rest will be covered in the course.

  • You are not required to know anything about forecasting - we will teach you;
  • You are not expected to know statistics - we will explain the essentials;
  • You do not need to be a programming language expert - we will show you how to become one;
About the Centre
Centre for Marketing Analytics and Forecasting, Lancaster University
Marketing Analytics
Supply Chain Forecasting
S&OP Process
Machine Learning
Software Development
Inventory Management

Our Centre, founded in 1990, continues to lead the field in applied forecasting and marketing analytics. Our research, executive support and training helps practitioners and academics working in the retail, manufacturing, telecommunications and software sectors. We develop new methods to tackle a range of problems facing those using predictive analytics, including demand planning and marketing modelling. Our work results in substantial cost reductions and service level improvements for a range of private and public organisations.

Specifically, we get involved with projects for multinational companies and we advise companies on effective procedures for forecasting and inventory management. We also recommend and design the best software solutions for the operational side of businesses. In addition to this we design and deliver courses to meet specific group needs, using interactive material that gives attendees hands-on experience in producing effective forecasts. We are the only centre in the UK to offer a Certificate of Forecasting issued by the International Institute of Forecasters (IIF).

Throughout the year we host a range of events, including courses and guest speakers, it is a great way to tap into our expertise. And our list of publications demonstrates the scope of our work.

Questions and Answers

One week.

Every session lasts for 3 hours with several breaks, giving opportunity to communicate with tutors and ask additional questions.

Yes, in order to attend, you need to register and pay for the course.

The course will be delivered in English. All supplementary materials are also in English.

 

German, Spanish, Russian and Greek can be considered as alternative options for the bespoke course on a similar topic, but need to be discussed separately with the tutors. Get in touch with us on cmaf@lancaster.ac.uk

The course is conducted online, so it does not matter as long as you understand English and can join the online sessions.

So, are you ready to become a specialist in forecasting?

Then what are you waiting for? Sign up today!

CMAF on social media