To Accompany Business Statistics A Decision Making Approach 8th Ed Chapter 16 Analyzing and Forecasting TimeSeries Data By Groebner Shannon Fry amp Smith PrenticeHall Publishing Company ID: 426123
Download Presentation The PPT/PDF document "Guide to Using Minitab 14 For Basic Stat..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Guide to Using Minitab 14 For Basic Statistical Applications
To AccompanyBusiness Statistics: A Decision Making Approach, 8th Ed.Chapter 16:Analyzing and Forecasting Time-Series DataByGroebner, Shannon, Fry, & SmithPrentice-Hall Publishing CompanyCopyright, 2011Slide2
Chapter 16 Minitab ExamplesTrend Based Forecasting
Taft Ice Cream CompanyNonlinear Trend Harrison Equipment CompanySeasonal Adjustment Big Mountain Ski ResortSingle Exponential Smoothing Dawson Graphic DesignsMore ExamplesSlide3
Chapter 16 Minitab ExamplesDouble Exponential Smoothing
Billingsley Insurance CompanySlide4
Trend Based Forecasting - Taft Ice Cream Company
Issue: The owners of Taft Ice Cream Company considering expanding their manufacturing facilities. The bank requires a forecast of future sales. Objective: Use Minitab to build a forecasting model based on 10 years of data. Data file is Taft.MTWSlide5
Open File
Taft.MTWTrend Based Forecasting – Taft Ice Cream CompanySlide6
First click on
Graph, then Time Series Plot.Trend Based Forecasting – Taft Ice Cream CompanySlide7
Select
SimpleTrend Based Forecasting – Taft Ice Cream CompanySlide8
Enter
data column to be graphed. Then click Time/ScaleTrend Based Forecasting – Taft Ice Cream CompanySlide9
Click on
Calendar and specify Year – then determine starting Year (1997)Trend Based Forecasting – Taft Ice Cream CompanySlide10
A linear trend is evident in this time series plot.
Trend Based Forecasting – Taft Ice Cream CompanySlide11
To develop the trend line, click on
Stat, then Regression and Regression againTrend Based Forecasting – Taft Ice Cream CompanySlide12
Identify the columns containing the
Time Series and also specify the dependent variable (t) Trend Based Forecasting – Taft Ice Cream CompanySlide13
Linear Trend Model
Trend Based Forecasting – Taft Ice Cream CompanySlide14
A second method - select
Stat – Time Series – Trend AnalysisTrend Based Forecasting – Taft Ice Cream CompanySlide15
Specify time series variable and select
Linear model typeTrend Based Forecasting – Taft Ice Cream CompanySlide16
Measures of forecast accuracy
Trend Based Forecasting – Taft Ice Cream CompanySlide17
Issue:
Harrison Equipment is interested in forecasting future repair costs for a crawler tractor it leases to contractors. Objective: Use Minitab to develop a nonlinear forecasting model. Data file is Harrison .MTWNonlinear Trend – Harrison Equipment CompanySlide18
Open File
Harrison.MTWNonlinear Trend – Harrison Equipment CompanySlide19
Select
Graph – then Time Series PlotNonlinear Trend – Harrison Equipment CompanySlide20
Select
SimpleNonlinear Trend – Harrison Equipment CompanySlide21
Define Time Series variable (Repair Costs) and then select
Time/ScaleNonlinear Trend – Harrison Equipment CompanySlide22
Select
Calendar and pick Quarter Year optionNonlinear Trend – Harrison Equipment CompanySpecify starting
Quarter
and
YearSlide23
Nonlinear Trend – Harrison Equipment CompanySlide24
To develop a linear model, click on
Stat, then Time Series and finally Trend Analysis.Nonlinear Trend – Harrison Equipment CompanySlide25
Specify time series variable (Repair Costs) and select
LinearNonlinear Trend – Harrison Equipment CompanySlide26
Linear Model Results
Nonlinear Trend – Harrison Equipment CompanySlide27
To develop a model with time squared as the variable Click on
Calc, then Calculator.Nonlinear Trend – Harrison Equipment CompanySlide28
Identify column for new variable, in Expressions box enter form of new variable. Click
OKNonlinear Trend – Harrison Equipment CompanySlide29
Click on
Stat, then Regression and Regression again.Nonlinear Trend – Harrison Equipment CompanySlide30
Define the Response variable (
Repair Costs) Predictors (Qtr2) then click Storage.Nonlinear Trend – Harrison Equipment CompanySlide31
Under Diagnostic Measures select Residuals, under Characteristics select Fits
. Click OK twice.Nonlinear Trend – Harrison Equipment CompanySlide32
The Minitab output shows the regression model.
Nonlinear Trend – Harrison Equipment CompanySlide33
Seasonal Adjustment -
Big Mountain Ski ResortIssue: The resort wants to build a forecasting model from data that has a definite seasonal component.Objective: Use Minitab to develop a forecasting model adjusting for seasonal data. Data file is Big Mountain.MTWSlide34
Open File
Big Mountain.MTWSeasonal Adjustment – Big Mountain Ski ResortSlide35
Seasonal Adjustment – Big Mountain Ski Resort
Click on Stat
,
then
Time Series
and then select
Decomposition
. Slide36
Define the
Variable, the Model Type, the Seasonal length and the Model Components.
Seasonal Adjustment – Big Mountain Ski ResortSlide37
The graph shows the actual and predicted values.
Seasonal Adjustment – Big Mountain Ski ResortSlide38
This output shows the original data and other graphs.
Seasonal Adjustment – Big Mountain Ski ResortSlide39
The
Trend Line Equation, the Seasonal Indices and MAPE, MAD and MSD are also given.Seasonal Adjustment – Big Mountain Ski ResortSlide40
Single Exponential Smoothing Dawson Graphic Design
Issue: The company needs to develop a forecasting model to forecast incoming customer calls so they are able to make informed future staffing decisions. Because the time series appears to be relatively stable, a relatively small smoothing constant will be used.Objective: Use Minitab to develop a single exponential smoothing forecasting model. Data file is Dawson.MTWSlide41
Open File
Dawson.MTWSingle Exponential Smoothing – Dawson Graphic DesignSlide42
Click on
Stat, then Time Series and finally Single Exponential Smoothing.Single Exponential Smoothing – Dawson Graphic DesignSlide43
Identify the
Time Series Variable. Either specify alpha or ask Minitab to optimize the forecasting model. Select StorageSingle Exponential Smoothing – Dawson Graphic DesignSlide44
Select
FitsSingle Exponential Smoothing – Dawson Graphic DesignSlide45
The graph shows the actual and forecast values. The
accuracy measures are also given.Single Exponential Smoothing – Dawson Graphic DesignSlide46
To determine optimal alpha, Identify the
Time Series Variable. Ask Minitab to optimize the forecasting model.Single Exponential Smoothing – Dawson Graphic DesignSlide47
Single Exponential Smoothing – Dawson Graphic Design
The graph shows the actual and forecast values. The accuracy measures and the optimum alpha are also given.Slide48
Issue:
The claims manager has data for 12 months and wants to forecast claims for month 13. But the time series contains a strong upward trend Objective: Use Minitab to develop a double exponential smoothing model. Data file is Billingsley.MTWDouble Exponential Smoothing Billingsley InsuranceSlide49
Open file
Billingsley.MTWDouble Exponential Smoothing – Billingsley InsuranceSlide50
Click on
Stat then Time Series and finally Double Exponential Smoothing. Double Exponential Smoothing – Billingsley InsuranceSlide51
Identify the
Time Series Variable. Either specify alpha and beta or ask Minitab to optimize the forecasting model.Double Exponential Smoothing – Billingsley InsuranceSlide52
The graph shows the actual and forecast values. The
accuracy measures are also given.Double Exponential Smoothing – Billingsley Insurance