Scribd Logo
Upload

Welcome to Scribd!

  • 190 views

    Assignment 2 - 2020

    Uploaded by

    Mpho Nku
    This document contains an assignment with two questions regarding demand forecasting. Question 1 involves analyzing the relationship between TV appearances and soft drink demand, calculating a regression line, and interpreting the r-squared value. Question 2 asks the student to forecast demand for brandy using several techniques, including moving averages and exponential smoothing, and determine the most accurate technique based on MAD. Formulas for calculations are provided.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Assignment 2 - 2020

    Uploaded by

    Mpho Nku
    190 views3 pages
    This document contains an assignment with two questions regarding demand forecasting. Question 1 involves analyzing the relationship between TV appearances and soft drink demand, calculating a regression line, and interpreting the r-squared value. Question 2 asks the student to forecast demand for brandy using several techniques, including moving averages and exponential smoothing, and determine the most accurate technique based on MAD. Formulas for calculations are provided.

    Original Description:

    Quantitative Analysis

    Original Title

    Assignment 2 - 2020 (4)

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document contains an assignment with two questions regarding demand forecasting. Question 1 involves analyzing the relationship between TV appearances and soft drink demand, calculating a regression line, and interpreting the r-squared value. Question 2 asks the student to forecast demand for brandy using several techniques, including moving averages and exponential smoothing, and determine the most accurate technique based on MAD. Formulas for calculations are provided.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    190 views3 pages

    Assignment 2 - 2020

    Uploaded by

    Mpho Nku
    This document contains an assignment with two questions regarding demand forecasting. Question 1 involves analyzing the relationship between TV appearances and soft drink demand, calculating a regression line, and interpreting the r-squared value. Question 2 asks the student to forecast demand for brandy using several techniques, including moving averages and exponential smoothing, and determine the most accurate technique based on MAD. Formulas for calculations are provided.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 3

    Assignment 2 (75 Marks)

    Question 1 ( 25 Marks )

    As a manager of a soft drink supplier, you want to determine if the appearance of your
    particular soft drink brand on TV are related to the demand for your soft drink. Depending
    on the relationship you may consider to continue use TV as a marketing platform. The data
    you collected are represented in the following table:

    Demand for Number of


    soft drink appearances on TV
    (Y) (X)
    3 3
    6 4
    7 7
    5 6
    10 8
    8 5

    a) Draw a graph showing the relationship between the number of appearances of


    your soft drink and the demand thereof. (4)
    b) Describe the relationship in your opinion? (2)
    c) Calculate the least square regression line for your data. Show all calculations in
    a tabular format. Use the formulas in appendix for calculations. (12)
    d) What is the expected estimate demand for your soft drinks if it appears 20 times
    on TV?
    In your opinion, is your demand estimate representative of reality? (3)
    e) Give the values for r2 and r for your model.
    How will you interpret the value of r2 for your problem? (4)

    CMPG322 1/3
    Question 2 ( 50 Marks )

    The table below represent a yearly demand for liters of brandy expressed in 1000’s
    collected by you for a local Liquor distributor in the Vaal triangle for 2019. You as the
    manager for this Liquor distributor want to improve your forecast accuracy for the next year
    (year 12). Being a student of CMPG322 you want to apply forecasting techniques, you
    have studied, to your data to make your forecast effort based on a scientific approach
    rather than haphazard forecasting approaches. You want to determine which of the
    selected forecasting techniques is the most accurate based on the MAD decision criteria.

    Year Demand
    (1000’s)
    1 4
    2 6
    3 4
    4 5
    5 10
    6 8
    7 7
    8 9
    9 12
    10 14
    11 15

    You chose the following forecasting techniques to estimate / forecast sales:

    a) A three year moving average. (10)

    b) A weighted moving average using a weight of 2 for the most recent year to your
    estimation and a weight of 1 for both years prior to the most recent year. (11)
    (i.e : estimation for year 8 = demand for year 7 (weight = 2), demand for years 6 & 5
    (weight =1).

    c) Used Y = 2.2181 +1.0545(x) as a trend line to estimate the yearly demand. (13)
    Where (x = time period (1, 2, . . ., 11); Y = demand),
    d) An exponential smoothing forecast using a smoothing constant of 0.3. Assume that
    the last period’s forecast for year 1 is 5000 liters of brandy. (14)

    e) Which forecasting technique is the best by making use of the MAD decision criteria
    for the above techniques (a – d). Explain why. (2)

    Total : 75

    CMPG322 2/3
    FORMULAS/FORMULES

    Y^ = bo +b1 X Y^ = bo +b1 X 1 +b 2 X 2 +. . .+bk X k

    b1=∑ (X - { X̄ )(Y - { Ȳ ) b0 =Ȳ - b 1 X̄


    ¿¿¿
    ∑ (X- { X̄ )2
    { X̄ =
    ∑X¿ { Ȳ =
    ∑Y ¿
    n n
    2 2
    SST=∑ (Y −Ȳ ) SSE=∑ (Y −Y^ )
    2
    SSR=∑ ( Y^ −Ȳ ) SST = SSR + SSE

    r=± √r 2 SSR SSE


    r 2= or r2=1−
    SST SST
    ∑ ( sum of demands in previous n periods ) Bias =
    ∑ error
    Moving average= n
    n
    ∑ ( weight for period n )( demand in period n) MAD =
    ∑|error|
    Weighted moving average=
    ∑ weights n

    Ft+1 = Ft + (Yt – Ft) e rror =A ctual value−Forecast value

    CMPG322 3/3

    You might also like