forecasting - rob j hyndman · fertilizer sales. brief bio forecasting: principles and practice...
TRANSCRIPT
![Page 1: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/1.jpg)
1. Introduction to forecasting
OTexts.org/fpp/1/OTexts.org/fpp/2/3
Forecasting: Principles and Practice 1
Rob J Hyndman
Forecasting:
Principles and Practice
![Page 2: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/2.jpg)
Resources
Slides
Exercises
Textbook
Useful links
robjhyndman.com/uwa
Forecasting: Principles and Practice 2
![Page 3: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/3.jpg)
Outline
1 Introduction
2 Some case studies
3 Time series data
4 Some simple forecasting methods
Forecasting: Principles and Practice Introduction 3
![Page 4: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/4.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 5: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/5.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 6: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/6.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 7: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/7.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 8: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/8.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 9: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/9.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 10: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/10.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 11: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/11.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer sales
![Page 12: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/12.jpg)
Brief bio
Forecasting: Principles and Practice Introduction 4
Director of Monash University’sBusiness & EconomicForecasting Unit
Editor-in-Chief, InternationalJournal of Forecasting
How my forecasting methodology is used:
Pharmaceutical Benefits Scheme
Cancer incidence and mortality
Electricity demand
Ageing population
Fertilizer salesrobjhyndman.com
![Page 13: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/13.jpg)
Poll: How experienced are you in forecasting?
1 Guru: I wrote the book, done it for decades,now I do the conference circuit.
2 Expert: It has been my full time job for morethan a decade.
3 Skilled: I have been doing it for years.
4 Comfortable: I understand it and have done it.
5 Learner: I am still learning.
6 Beginner: I have heard of it and would like tolearn more.
7 Unknown: What is forecasting? Is that what theweather people do?
Forecasting: Principles and Practice Introduction 5
![Page 14: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/14.jpg)
Key reference
Hyndman, R. J. & Athanasopoulos, G.
(2013) Forecasting: principles and
practice.
otexts.org/fpp/
Free and online
Data sets in associated R package
R code for examples
Forecasting: Principles and Practice Introduction 6
![Page 15: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/15.jpg)
Key reference
Hyndman, R. J. & Athanasopoulos, G.
(2013) Forecasting: principles and
practice.
otexts.org/fpp/
Free and online
Data sets in associated R package
R code for examples
Forecasting: Principles and Practice Introduction 6
![Page 16: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/16.jpg)
Key reference
Hyndman, R. J. & Athanasopoulos, G.
(2013) Forecasting: principles and
practice.
otexts.org/fpp/
Free and online
Data sets in associated R package
R code for examples
Forecasting: Principles and Practice Introduction 6
![Page 17: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/17.jpg)
Key reference
Hyndman, R. J. & Athanasopoulos, G.
(2013) Forecasting: principles and
practice.
otexts.org/fpp/
Free and online
Data sets in associated R package
R code for examples
Forecasting: Principles and Practice Introduction 6
![Page 18: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/18.jpg)
Key reference
Hyndman, R. J. & Athanasopoulos, G.
(2013) Forecasting: principles and
practice.
otexts.org/fpp/
Free and online
Data sets in associated R package
R code for examples
Forecasting: Principles and Practice Introduction 6
![Page 19: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/19.jpg)
Poll: How proficient are you in using R?
1 Guru: The R core team come to me for advice.2 Expert: I have written several packages on
CRAN.3 Skilled: I use it regularly and it is an important
part of my job.4 Comfortable: I use it often and am comfortable
with the tool.5 User: I use it sometimes, but I am often
searching around for the right function.6 Learner: I have used it a few times.7 Beginner: I’ve managed to download and install
it.8 Unknown: Why are you speaking like a pirate?
Forecasting: Principles and Practice Introduction 7
![Page 20: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/20.jpg)
Which version of R are you using?
Version: (try getRversion() if you don’t know)
1 R 3.0.0 or higher
2 R 2.15.x
3 R 2.14.x
4 Something older.
Edition
1 Standard R (CRAN)
2 Standard R with RStudio
3 Revolution R: Community, EnterpriseWorkstation or Server
4 Something else?Forecasting: Principles and Practice Introduction 8
![Page 21: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/21.jpg)
Install required packages
install.packages("fpp", dependencies=TRUE)
Forecasting: Principles and Practice Introduction 9
![Page 22: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/22.jpg)
Getting help with R# Search for termshelp.search("forecasting")
# Detailed helphelp(forecast)
# Worked examplesexample("forecast.ar")
# Similar namesapropos("forecast")
#Help on packagehelp(package="fpp")
Forecasting: Principles and Practice Introduction 10
![Page 23: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/23.jpg)
Approximate outlineDay Topic Chapter
1 The forecaster’s toolbox 1,21 Seasonality and trends 61 Exponential smoothing 7
2 Time series decomposition 62 Time series cross-validation 22 Transformations 22 Stationarity and differencing 82 ARIMA models 8
3 State space models –3 Dynamic regression 93 Hierarchical forecasting 93 Advanced methods 9
Forecasting: Principles and Practice Introduction 11
![Page 24: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/24.jpg)
Assumptions
This is not an introduction to R. I assume youare broadly comfortable with R code and the Renvironment.
This is not a statistics course. I assume you arefamiliar with concepts such as the mean,standard deviation, quantiles, regression,normal distribution, etc.
This is not a theory course. I am not going toderive anything. I will teach you forecastingtools, when to use them and how to use themmost effectively.
Forecasting: Principles and Practice Introduction 12
![Page 25: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/25.jpg)
Assumptions
This is not an introduction to R. I assume youare broadly comfortable with R code and the Renvironment.
This is not a statistics course. I assume you arefamiliar with concepts such as the mean,standard deviation, quantiles, regression,normal distribution, etc.
This is not a theory course. I am not going toderive anything. I will teach you forecastingtools, when to use them and how to use themmost effectively.
Forecasting: Principles and Practice Introduction 12
![Page 26: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/26.jpg)
Assumptions
This is not an introduction to R. I assume youare broadly comfortable with R code and the Renvironment.
This is not a statistics course. I assume you arefamiliar with concepts such as the mean,standard deviation, quantiles, regression,normal distribution, etc.
This is not a theory course. I am not going toderive anything. I will teach you forecastingtools, when to use them and how to use themmost effectively.
Forecasting: Principles and Practice Introduction 12
![Page 27: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/27.jpg)
Outline
1 Introduction
2 Some case studies
3 Time series data
4 Some simple forecasting methods
Forecasting: Principles and Practice Some case studies 13
![Page 28: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/28.jpg)
CASE STUDY 1: Paperware company
Forecasting: Principles and Practice Some case studies 14
Problem: Want forecasts of each ofhundreds of items. Series can bestationary, trended or seasonal. Theycurrently have a large forecastingprogram written in-house but it doesn’tseem to produce sensible forecasts.They want me to tell them what iswrong and fix it.
![Page 29: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/29.jpg)
CASE STUDY 1: Paperware company
Forecasting: Principles and Practice Some case studies 14
Problem: Want forecasts of each ofhundreds of items. Series can bestationary, trended or seasonal. Theycurrently have a large forecastingprogram written in-house but it doesn’tseem to produce sensible forecasts.They want me to tell them what iswrong and fix it.
Additional informationProgram written in COBOL making numerical calculationslimited. It is not possible to do any optimisation.
![Page 30: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/30.jpg)
CASE STUDY 1: Paperware company
Forecasting: Principles and Practice Some case studies 14
Problem: Want forecasts of each ofhundreds of items. Series can bestationary, trended or seasonal. Theycurrently have a large forecastingprogram written in-house but it doesn’tseem to produce sensible forecasts.They want me to tell them what iswrong and fix it.
Additional informationProgram written in COBOL making numerical calculationslimited. It is not possible to do any optimisation.Their programmer has little experience in numericalcomputing.
![Page 31: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/31.jpg)
CASE STUDY 1: Paperware company
Forecasting: Principles and Practice Some case studies 14
Problem: Want forecasts of each ofhundreds of items. Series can bestationary, trended or seasonal. Theycurrently have a large forecastingprogram written in-house but it doesn’tseem to produce sensible forecasts.They want me to tell them what iswrong and fix it.
Additional informationProgram written in COBOL making numerical calculationslimited. It is not possible to do any optimisation.Their programmer has little experience in numericalcomputing.They employ no statisticians and want the program toproduce forecasts automatically.
![Page 32: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/32.jpg)
CASE STUDY 1: Paperware company
Methods currently used
A 12 month average
C 6 month average
E straight line regression over last 12 months
G straight line regression over last 6 months
H average slope between last year’s and thisyear’s values.(Equivalent to differencing at lag 12 andtaking mean.)
I Same as H except over 6 months.
K I couldn’t understand the explanation.
Forecasting: Principles and Practice Some case studies 15
![Page 33: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/33.jpg)
CASE STUDY 2: PBS
Forecasting: Principles and Practice Some case studies 16
![Page 34: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/34.jpg)
CASE STUDY 2: PBS
The Pharmaceutical Benefits Scheme (PBS) isthe Australian government drugs subsidy scheme.
Many drugs bought from pharmacies aresubsidised to allow more equitable access tomodern drugs.
The cost to government is determined by thenumber and types of drugs purchased.Currently nearly 1% of GDP.
The total cost is budgeted based on forecastsof drug usage.
Forecasting: Principles and Practice Some case studies 17
![Page 35: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/35.jpg)
CASE STUDY 2: PBS
The Pharmaceutical Benefits Scheme (PBS) isthe Australian government drugs subsidy scheme.
Many drugs bought from pharmacies aresubsidised to allow more equitable access tomodern drugs.
The cost to government is determined by thenumber and types of drugs purchased.Currently nearly 1% of GDP.
The total cost is budgeted based on forecastsof drug usage.
Forecasting: Principles and Practice Some case studies 17
![Page 36: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/36.jpg)
CASE STUDY 2: PBS
The Pharmaceutical Benefits Scheme (PBS) isthe Australian government drugs subsidy scheme.
Many drugs bought from pharmacies aresubsidised to allow more equitable access tomodern drugs.
The cost to government is determined by thenumber and types of drugs purchased.Currently nearly 1% of GDP.
The total cost is budgeted based on forecastsof drug usage.
Forecasting: Principles and Practice Some case studies 17
![Page 37: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/37.jpg)
CASE STUDY 2: PBS
The Pharmaceutical Benefits Scheme (PBS) isthe Australian government drugs subsidy scheme.
Many drugs bought from pharmacies aresubsidised to allow more equitable access tomodern drugs.
The cost to government is determined by thenumber and types of drugs purchased.Currently nearly 1% of GDP.
The total cost is budgeted based on forecastsof drug usage.
Forecasting: Principles and Practice Some case studies 17
![Page 38: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/38.jpg)
CASE STUDY 2: PBS
Forecasting: Principles and Practice Some case studies 18
![Page 39: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/39.jpg)
CASE STUDY 2: PBS
In 2001: $4.5 billion budget, under-forecastedby $800 million.Thousands of products. Seasonal demand.Subject to covert marketing, volatile products,uncontrollable expenditure.Although monthly data available for 10 years,data are aggregated to annual values, and onlythe first three years are used in estimating theforecasts.All forecasts being done with the FORECASTfunction in MS-Excel!
Problem: How to do the forecasting better?
Forecasting: Principles and Practice Some case studies 19
![Page 40: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/40.jpg)
CASE STUDY 2: PBS
In 2001: $4.5 billion budget, under-forecastedby $800 million.Thousands of products. Seasonal demand.Subject to covert marketing, volatile products,uncontrollable expenditure.Although monthly data available for 10 years,data are aggregated to annual values, and onlythe first three years are used in estimating theforecasts.All forecasts being done with the FORECASTfunction in MS-Excel!
Problem: How to do the forecasting better?
Forecasting: Principles and Practice Some case studies 19
![Page 41: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/41.jpg)
CASE STUDY 2: PBS
In 2001: $4.5 billion budget, under-forecastedby $800 million.Thousands of products. Seasonal demand.Subject to covert marketing, volatile products,uncontrollable expenditure.Although monthly data available for 10 years,data are aggregated to annual values, and onlythe first three years are used in estimating theforecasts.All forecasts being done with the FORECASTfunction in MS-Excel!
Problem: How to do the forecasting better?
Forecasting: Principles and Practice Some case studies 19
![Page 42: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/42.jpg)
CASE STUDY 2: PBS
In 2001: $4.5 billion budget, under-forecastedby $800 million.Thousands of products. Seasonal demand.Subject to covert marketing, volatile products,uncontrollable expenditure.Although monthly data available for 10 years,data are aggregated to annual values, and onlythe first three years are used in estimating theforecasts.All forecasts being done with the FORECASTfunction in MS-Excel!
Problem: How to do the forecasting better?
Forecasting: Principles and Practice Some case studies 19
![Page 43: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/43.jpg)
CASE STUDY 2: PBS
In 2001: $4.5 billion budget, under-forecastedby $800 million.Thousands of products. Seasonal demand.Subject to covert marketing, volatile products,uncontrollable expenditure.Although monthly data available for 10 years,data are aggregated to annual values, and onlythe first three years are used in estimating theforecasts.All forecasts being done with the FORECASTfunction in MS-Excel!
Problem: How to do the forecasting better?
Forecasting: Principles and Practice Some case studies 19
![Page 44: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/44.jpg)
CASE STUDY 2: PBS
In 2001: $4.5 billion budget, under-forecastedby $800 million.Thousands of products. Seasonal demand.Subject to covert marketing, volatile products,uncontrollable expenditure.Although monthly data available for 10 years,data are aggregated to annual values, and onlythe first three years are used in estimating theforecasts.All forecasts being done with the FORECASTfunction in MS-Excel!
Problem: How to do the forecasting better?
Forecasting: Principles and Practice Some case studies 19
![Page 45: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/45.jpg)
CASE STUDY 3: Airline
Forecasting: Principles and Practice Some case studies 20
![Page 46: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/46.jpg)
CASE STUDY 3: Airline
Forecasting: Principles and Practice Some case studies 21
First class passengers: Melbourne−Sydney
Year
1988 1989 1990 1991 1992 1993
0.0
1.0
2.0
Business class passengers: Melbourne−Sydney
Year
1988 1989 1990 1991 1992 1993
02
46
8
Economy class passengers: Melbourne−Sydney
Year
1988 1989 1990 1991 1992 1993
010
2030
![Page 47: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/47.jpg)
CASE STUDY 3: Airline
Forecasting: Principles and Practice Some case studies 21
First class passengers: Melbourne−Sydney
Year
1988 1989 1990 1991 1992 1993
0.0
1.0
2.0
Business class passengers: Melbourne−Sydney
Year
1988 1989 1990 1991 1992 1993
02
46
8
Economy class passengers: Melbourne−Sydney
Year
1988 1989 1990 1991 1992 1993
010
2030
Not the real data!Or is it?
![Page 48: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/48.jpg)
CASE STUDY 3: Airline
Problem: how to forecast passenger traffic onmajor routes.
Additional information
They can provide a large amount of data onprevious routes.
Traffic is affected by school holidays, specialevents such as the Grand Prix, advertisingcampaigns, competition behaviour, etc.
They have a highly capable team of people whoare able to do most of the computing.
Forecasting: Principles and Practice Some case studies 22
![Page 49: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/49.jpg)
CASE STUDY 3: Airline
Problem: how to forecast passenger traffic onmajor routes.
Additional information
They can provide a large amount of data onprevious routes.
Traffic is affected by school holidays, specialevents such as the Grand Prix, advertisingcampaigns, competition behaviour, etc.
They have a highly capable team of people whoare able to do most of the computing.
Forecasting: Principles and Practice Some case studies 22
![Page 50: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/50.jpg)
CASE STUDY 3: Airline
Problem: how to forecast passenger traffic onmajor routes.
Additional information
They can provide a large amount of data onprevious routes.
Traffic is affected by school holidays, specialevents such as the Grand Prix, advertisingcampaigns, competition behaviour, etc.
They have a highly capable team of people whoare able to do most of the computing.
Forecasting: Principles and Practice Some case studies 22
![Page 51: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/51.jpg)
CASE STUDY 3: Airline
Problem: how to forecast passenger traffic onmajor routes.
Additional information
They can provide a large amount of data onprevious routes.
Traffic is affected by school holidays, specialevents such as the Grand Prix, advertisingcampaigns, competition behaviour, etc.
They have a highly capable team of people whoare able to do most of the computing.
Forecasting: Principles and Practice Some case studies 22
![Page 52: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/52.jpg)
Outline
1 Introduction
2 Some case studies
3 Time series data
4 Some simple forecasting methods
Forecasting: Principles and Practice Time series data 23
![Page 53: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/53.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 54: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/54.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 55: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/55.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 56: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/56.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 57: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/57.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 58: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/58.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 59: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/59.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 60: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/60.jpg)
Time series data
Time series consist of sequences ofobservations collected over time.We will assume the time periods are equallyspaced.
Time series examples
Daily IBM stock pricesMonthly rainfallAnnual Google profitsQuarterly Australian beer production
Forecasting is estimating how the sequenceof observations will continue into the future.
Forecasting: Principles and Practice Time series data 24
![Page 61: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/61.jpg)
Australian beer production
Forecasting: Principles and Practice Time series data 25
Year
meg
alite
rs
1995 2000 2005 2010
400
450
500
![Page 62: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/62.jpg)
Looking for stories
Forecasting: Principles and Practice Time series data 26
![Page 63: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/63.jpg)
Looking for stories that make sense
Forecasting: Principles and Practice Time series data 27
![Page 64: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/64.jpg)
Think about what you’re doing
Forecasting: Principles and Practice Time series data 28
![Page 65: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/65.jpg)
Time series in RAustralian GDPausgdp <- ts(scan("gdp.dat"),frequency=4,
start=1971+2/4)Class: tsPrint and plotting methods available.
> ausgdpQtr1 Qtr2 Qtr3 Qtr4
1971 4612 46511972 4645 4615 4645 47221973 4780 4830 4887 49331974 4921 4875 4867 49051975 4938 4934 4942 49791976 5028 5079 5112 51271977 5130 5101 5072 50691978 5100 5166 5244 53121979 5349 5370 5388 53961980 5388 5403 5442 5482
Forecasting: Principles and Practice Time series data 29
![Page 66: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/66.jpg)
Time series in RAustralian GDPausgdp <- ts(scan("gdp.dat"),frequency=4,
start=1971+2/4)Class: tsPrint and plotting methods available.
> ausgdpQtr1 Qtr2 Qtr3 Qtr4
1971 4612 46511972 4645 4615 4645 47221973 4780 4830 4887 49331974 4921 4875 4867 49051975 4938 4934 4942 49791976 5028 5079 5112 51271977 5130 5101 5072 50691978 5100 5166 5244 53121979 5349 5370 5388 53961980 5388 5403 5442 5482
Forecasting: Principles and Practice Time series data 29
![Page 67: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/67.jpg)
Time series in RAustralian GDPausgdp <- ts(scan("gdp.dat"),frequency=4,
start=1971+2/4)Class: tsPrint and plotting methods available.
> ausgdpQtr1 Qtr2 Qtr3 Qtr4
1971 4612 46511972 4645 4615 4645 47221973 4780 4830 4887 49331974 4921 4875 4867 49051975 4938 4934 4942 49791976 5028 5079 5112 51271977 5130 5101 5072 50691978 5100 5166 5244 53121979 5349 5370 5388 53961980 5388 5403 5442 5482
Forecasting: Principles and Practice Time series data 29
![Page 68: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/68.jpg)
Time series in R
Forecasting: Principles and Practice Time series data 30
Time
ausg
dp
1975 1980 1985 1990 19954500
5000
5500
6000
6500
7000
7500 > plot(ausgdp)
![Page 69: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/69.jpg)
Time series in R
Residential electricity sales
> elecsalesTime Series:Start = 1989End = 2008Frequency = 1[1] 2354.34 2379.71 2318.52 2468.99 2386.09 2569.47[7] 2575.72 2762.72 2844.50 3000.70 3108.10 3357.50
[13] 3075.70 3180.60 3221.60 3176.20 3430.60 3527.48[19] 3637.89 3655.00
Forecasting: Principles and Practice Time series data 31
![Page 70: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/70.jpg)
Time series in RMain package used in this course> library(fpp)
Forecasting: Principles and Practice Time series data 32
![Page 71: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/71.jpg)
Time series in RMain package used in this course> library(fpp)This loads:
some data for use in examples and exercisesforecast package (for forecasting functions)tseries package (for a few time seriesfunctions)fma package (for lots of time series data)expsmooth package (for more time seriesdata)lmtest package (for some regressionfunctions)
Forecasting: Principles and Practice Time series data 32
![Page 72: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/72.jpg)
Time series in RMain package used in this course> library(fpp)This loads:
some data for use in examples and exercisesforecast package (for forecasting functions)tseries package (for a few time seriesfunctions)fma package (for lots of time series data)expsmooth package (for more time seriesdata)lmtest package (for some regressionfunctions)
Forecasting: Principles and Practice Time series data 32
![Page 73: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/73.jpg)
Time series in RMain package used in this course> library(fpp)This loads:
some data for use in examples and exercisesforecast package (for forecasting functions)tseries package (for a few time seriesfunctions)fma package (for lots of time series data)expsmooth package (for more time seriesdata)lmtest package (for some regressionfunctions)
Forecasting: Principles and Practice Time series data 32
![Page 74: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/74.jpg)
Time series in RMain package used in this course> library(fpp)This loads:
some data for use in examples and exercisesforecast package (for forecasting functions)tseries package (for a few time seriesfunctions)fma package (for lots of time series data)expsmooth package (for more time seriesdata)lmtest package (for some regressionfunctions)
Forecasting: Principles and Practice Time series data 32
![Page 75: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/75.jpg)
Time series in RMain package used in this course> library(fpp)This loads:
some data for use in examples and exercisesforecast package (for forecasting functions)tseries package (for a few time seriesfunctions)fma package (for lots of time series data)expsmooth package (for more time seriesdata)lmtest package (for some regressionfunctions)
Forecasting: Principles and Practice Time series data 32
![Page 76: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/76.jpg)
Time series in RMain package used in this course> library(fpp)This loads:
some data for use in examples and exercisesforecast package (for forecasting functions)tseries package (for a few time seriesfunctions)fma package (for lots of time series data)expsmooth package (for more time seriesdata)lmtest package (for some regressionfunctions)
Forecasting: Principles and Practice Time series data 32
![Page 77: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/77.jpg)
Outline
1 Introduction
2 Some case studies
3 Time series data
4 Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 33
![Page 78: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/78.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 34
Australian quarterly beer production
meg
alite
rs
1995 2000 2005
400
450
500
![Page 79: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/79.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 34
Australian quarterly beer production
meg
alite
rs
1995 2000 2005
400
450
500
Can you think of any forecasting methods for these data?
![Page 80: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/80.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 35
Number of pigs slaughtered in Victoria
thou
sand
s
1990 1991 1992 1993 1994 1995
8090
100
110
![Page 81: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/81.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 35
Number of pigs slaughtered in Victoria
thou
sand
s
1990 1991 1992 1993 1994 1995
8090
100
110
How would you forecast these data?
![Page 82: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/82.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 36
Dow Jones index (daily ending 15 Jul 94)
0 50 100 150 200 250
3600
3700
3800
3900
![Page 83: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/83.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 36
Dow Jones index (daily ending 15 Jul 94)
0 50 100 150 200 250
3600
3700
3800
3900
How would you forecast these data?
![Page 84: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/84.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 85: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/85.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 86: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/86.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 87: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/87.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 88: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/88.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 89: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/89.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 90: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/90.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 91: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/91.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 92: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/92.jpg)
Some simple forecasting methods
Average method
Forecast of all future values is equal to mean ofhistorical data {y1, . . . , yT}.Forecasts: yT+h|T = y = (y1 + · · ·+ yT)/T
Naïve method (for time series only)
Forecasts equal to last observed value.Forecasts: yT+h|T = yT.Consequence of efficient market hypothesis.
Seasonal naïve method
Forecasts equal to last value from same season.Forecasts: yT+h|T = yT+h−km where m =seasonal period and k = b(h− 1)/mc+1.
Forecasting: Principles and Practice Some simple forecasting methods 37
![Page 93: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/93.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 38
Forecasts for quarterly beer production
1995 2000 2005
400
450
500
Which method is which?
![Page 94: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/94.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 38
Forecasts for quarterly beer production
1995 2000 2005
400
450
500
Mean methodNaive methodSeasonal naive method
Which method is which?
![Page 95: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/95.jpg)
Drift method
Forecasts equal to last value plus averagechange.
Forecasts:
yT+h|T = yT +h
T − 1
T∑t=2
(yt − yt−1)
= yT +h
T − 1(yT − y1).
Equivalent to extrapolating a line drawnbetween first and last observations.
Forecasting: Principles and Practice Some simple forecasting methods 39
![Page 96: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/96.jpg)
Drift method
Forecasts equal to last value plus averagechange.
Forecasts:
yT+h|T = yT +h
T − 1
T∑t=2
(yt − yt−1)
= yT +h
T − 1(yT − y1).
Equivalent to extrapolating a line drawnbetween first and last observations.
Forecasting: Principles and Practice Some simple forecasting methods 39
![Page 97: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/97.jpg)
Drift method
Forecasts equal to last value plus averagechange.
Forecasts:
yT+h|T = yT +h
T − 1
T∑t=2
(yt − yt−1)
= yT +h
T − 1(yT − y1).
Equivalent to extrapolating a line drawnbetween first and last observations.
Forecasting: Principles and Practice Some simple forecasting methods 39
![Page 98: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/98.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 40
Dow Jones Index (daily ending 15 Jul 94)
Day
0 50 100 150 200 250 300
3600
3700
3800
3900
![Page 99: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/99.jpg)
Some simple forecasting methods
Forecasting: Principles and Practice Some simple forecasting methods 40
Dow Jones Index (daily ending 15 Jul 94)
Day
0 50 100 150 200 250 300
3600
3700
3800
3900
Mean methodNaive methodDrift model
![Page 100: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/100.jpg)
Some simple forecasting methods
Mean: meanf(x, h=20)
Naive: naive(x, h=20) or rwf(x, h=20)
Seasonal naive: snaive(x, h=20)
Drift: rwf(x, drift=TRUE, h=20)
Forecasting: Principles and Practice Some simple forecasting methods 41
![Page 101: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/101.jpg)
Some simple forecasting methods
Mean: meanf(x, h=20)
Naive: naive(x, h=20) or rwf(x, h=20)
Seasonal naive: snaive(x, h=20)
Drift: rwf(x, drift=TRUE, h=20)
Forecasting: Principles and Practice Some simple forecasting methods 41
![Page 102: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/102.jpg)
Some simple forecasting methods
Mean: meanf(x, h=20)
Naive: naive(x, h=20) or rwf(x, h=20)
Seasonal naive: snaive(x, h=20)
Drift: rwf(x, drift=TRUE, h=20)
Forecasting: Principles and Practice Some simple forecasting methods 41
![Page 103: Forecasting - Rob J Hyndman · Fertilizer sales. Brief bio Forecasting: Principles and Practice Introduction 4 ... CASE STUDY 2: PBS Forecasting: Principles and Practice Some case](https://reader034.vdocuments.mx/reader034/viewer/2022051308/5ae033027f8b9ab4688d04c0/html5/thumbnails/103.jpg)
Some simple forecasting methods
Mean: meanf(x, h=20)
Naive: naive(x, h=20) or rwf(x, h=20)
Seasonal naive: snaive(x, h=20)
Drift: rwf(x, drift=TRUE, h=20)
Forecasting: Principles and Practice Some simple forecasting methods 41