Scribd
Upload
28 views

Forecasting (II) PDF

The document discusses two methods for forecasting: exponential smoothing and regression analysis. Exponential smoothing weights recent observations more heavily than older observations using a smoothing constant between 0 and 1. Regression analysis uses a linear equation (Y=a+bX) to model the relationship between a dependent variable (Y) and independent variable (X). The document provides examples of using both methods to forecast output and sales for year 2021 based on historical data.

Uploaded by

Fausto Da Gama
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
28 views

Forecasting (II) PDF

The document discusses two methods for forecasting: exponential smoothing and regression analysis. Exponential smoothing weights recent observations more heavily than older observations using a smoothing constant between 0 and 1. Regression analysis uses a linear equation (Y=a+bX) to model the relationship between a dependent variable (Y) and independent variable (X). The document provides examples of using both methods to forecast output and sales for year 2021 based on historical data.

Uploaded by

Fausto Da Gama
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

ECONOMICS FOR ENGINEERS

Forecasting
Exponential Smoothing method
 Include all past observations
 Weight recent observations much more heavily than very old
observations
Exponential Smoothing method
 It uses a weighted average of past data as the basis for a
forecast
 The procedure gives heaviest weight to more recent
information and small weight to observations in the more
distant past.
 The formula for exponential smoothing is
 Y’t+1=  Yt+ (1- ) Y’t
 In other words
 Y’ 2021=Y 2020+(1-)Y’ 2020
 Where
 Y’t+1= Exponential smoothed average to be used as the
forecast
 Yt = Most recent actual data
 Y’t = Most recent smoothed forecast
  = smoothing constant

 Smoothing constant (Weight ) has a value between 0to 1


 The following table shows the actual output in last four years.
Using exponential smoothing , forecast the output for 2021,
where the smoothing constant is 0.4 and the starting forecast
is 23. Find predicted data for 2021

Year 2017 2018 2019 2020

Output 25 18 22 20
Regression analysis
 Y= a +b X
 Where Y= dependent variable
 X= Independent variable
 The following table shows actual output . Estimate sales for
the year 2021.
Year (X) 2012 2013 2014 2015 2016

Sales (Y) 120 130 135 142 138


 Y=a +bX

 b= Σ (X- x̄ ) (y - y̅ )
Σ (X- x̄ )2
= 4.8
 a = y̅ - b x̄
= 118.6
 Sales= 118.6+4.8 year
 Y’ sales= 118.6+4.8 (10)

Year (X) (X) Sales (Y)
2012 1 120
2013 2 130
2014 3 135
2015 4 142
2016 5 138
x̄ = 3 y̅= 133
Year (X) Sales (Y) (X- x̄ ) (y - y̅ ) (X- x̄ ) (y - y̅ ) (X- x̄ )2
(X)
2012 1 120 -2 -13 26 4
2013 2 130 -1 -3 3 1
2014 3 135 0 2 0 0
2015 4 142 1 9 9 1
2016 5 138 2 5 10 4
x̄ = ΣX y̅= ΣY Σ (X- x̄ ) (y - y̅ ) = Σ (X- x̄ )2
n n 48 = 10

You might also like