INTRODUCTION TO FORECASTING Chapter 1 – Getting Started
Nature and Uses of Forecasts
Forecasting problems occur in many fields: Business and industry Economics Finance Environmental sciences Social sciences Political sciences
Forecasting Problems Short-term forecasts Medium-term forecasts Predicting only a few periods ahead (hours, days, weeks) Typically bad on modeling and extrapolating patterns in the data Medium-term forecasts One to two years into the future, typically Long-term forecasts Several years into the future
Forecasting Problems Short-term forecasts Medium-term forecasts are needed for the scheduling of personnel, production and transportation. As part of the scheduling process, forecasts of demand are often also required. Medium-term forecasts are needed to determine future resource requirements, in order to purchase raw materials, hire personnel, or buy machinery and equipment. Long-term forecasts are used in strategic planning. Such decisions must take account of market opportunities, environmental factors and internal resources.
Most forecasting problems involve a time series:
Many business applications of forecasting utilize daily, weekly, monthly, quarterly, or annual data, but any reporting interval may be used. The data may be instantaneous, such as the viscosity of a chemical product at the point in time where it is measured; it may be cumulative, such as the total sales of a product during the month; or it may be a statistic that in some way reflects the activity of the variable during the time period, such as the daily closing price of a specific stock on the New York Stock Exchange.
The reason that forecasting is so important is that prediction of future events is a critical input into many types of planning and decision making processes, with application to areas such as the following: Operations Management. Business organizations routinely use forecasts of product sales or demand for services in order to schedule production, control inventories, manage the supply chain, determine staffing requirements, and plan capacity. Forecasts may also be used to determine the mix of products or services to be offered and the locations at which products are to be produced.
Introduction to Time Series Analysis and Forecasting, 2008 MJK Chapter 1 Introduction to Time Series Analysis and Forecasting, 2008 MJK
Two broad types of methods: Quantitative forecasting methods (focus of this course) Makes formal use of historical data A mathematical/statistical model Past patterns are modeled and projected into the future Qualitative forecasting methods (Chapter 4) Subjective Little available data (new product introduction) Expert opinion often used The Delphi method
Quantitative Forecasting Methods Graphical and exploratory methods Chapters 2 and 3 Regression methods Chapter 5 Smoothing methods Chapters 6 and 7 Formal time series analysis methods Chapters 8 and 9 Supervised learning methods (time permitting) Chapter 11 – neural networks, random forests, etc.
Terminology Point forecast or point estimate Single estimate of future values of the response in a time series, e.g. next month’s sales. Prediction or forecast interval Interval estimate for a future value of the response in a time series. This interval should have a high chance of covering the yet unobserved value, e.g. 80% or 95%. Forecast horizon or lead time How far out do we need forecasts for? e.g. next month, each month in the next year, next 4 years?
Examples of time series: Uncorrelated data, constant or stationary process model
Autocorrelated time series
Trend
Cyclic or seasonal time series
Nonstationary time series Note: The previous three examples were also nonstationary time series.
Another nonstationary time series
A mixture of patterns - nonstationary
Cyclic patterns of different magnitudes – again nonstationary
Atypical events or Anomalies
The Forecasting Process
The Forecasting Process Forecasting: Principles and Practice (5 steps)
The Forecasting Process Forecasting: Principles and Practice (5 steps)
Software There are numerous software packages that will allow us to analyze time series and make forecasts. R/R-Studio – open source! Constantly evolving with a huge user community. Lots of internet resources! JMP – SAS product has fairly substantial time series capabilities. We will use it some in this course. Others: MINITAB, SPSS, Stata, MATLAB, etc.
More Examples of Time Series – What features do you see?
More Examples of Time Series – What features do you see?
More Examples of Time Series – What features do you see? Monthly Sales Fastenal Corporation (01/01/04 – 12/31/2013)
More Examples of Time Series – What features do you see? Monthly U.S. Liquor Sales in Millions of Dollars (1980 -2007)
More Examples of Time Series – What features do you see? Original Time Series Plot of log10(Liquor Sales)
More Examples of Time Series – What features do you see? Original Time Series
For class next time: Read FPP Chapter 1 – Getting Started Download and install R from CRAN Download and install R-Studio Install R packages: forecast and fpp2 Work through all of the examples in the R Markdown file for Chapter 1 – Getting Started. Install JMP 13 Pro from the WSU network.