New Forecasting Techniques

download New Forecasting Techniques

of 46

Transcript of New Forecasting Techniques

  • 8/10/2019 New Forecasting Techniques

    1/46

    Dr. C. LightnerFayetteville State University

    1

    FORECASTING TECHNIQUES

    Chapter 16Qualitative Approaches to Forecasting

    Quantitative Approaches to Forecasting

    The Components of a Time Series

    Using Smoothing Methods in Forecasting

    Measures of Forecast Accuracy

    Using Trend Projection in Forecasting

    Using Regression Analysis in Forecasting

  • 8/10/2019 New Forecasting Techniques

    2/46

    Dr. C. LightnerFayetteville State University

    2

    Forecasting Introduction

    An essential aspect of managing any organization isplanning for the future.

    Organizations employ forecasting techniques to

    determine future inventory, costs, capacities, and

    interest rate changes.There are two basic approaches to forecasting:

    -Qualitative

    -Quantitative

  • 8/10/2019 New Forecasting Techniques

    3/46

    Dr. C. LightnerFayetteville State University

    3

    Qualitative Approaches to Forecasting

    Delphi Approach A panel of experts, each of whom is physically separated from

    the others and is anonymous, is asked to respond to a

    sequential series of questionnaires.

    After each questionnaire, the responses are tabulated and the

    information and opinions of the entire group are made known toeach of the other panel members so that they may revise their

    previous forecast response.

    The process continues until some degree of consensus is

    achieved.

  • 8/10/2019 New Forecasting Techniques

    4/46

    Dr. C. LightnerFayetteville State University

    4

    Qualitative Approaches (continued)

    Scenario Writing Scenario writing consists of developing a conceptual scenario

    of the future based on a well defined set of assumptions.

    After several different scenarios have been developed, the

    decision maker determines which is most likely to occur in the

    future and makes decisions accordingly.

  • 8/10/2019 New Forecasting Techniques

    5/46

    Dr. C. LightnerFayetteville State University

    5

    Qualitative Approaches (continued)

    Subjective or Interactive Approaches These techniques are often used by committees or panels

    seeking to develop new ideas or solve complex problems.

    They often involve "brainstorming sessions".

    It is important in such sessions that any ideas or opinions be

    permitted to be presented without regard to its relevancy and

    without fear of criticism.

  • 8/10/2019 New Forecasting Techniques

    6/46

    Dr. C. LightnerFayetteville State University

    6

    Quantitative Approaches to Forecasting

    Quantitative methods are based on an analysis of historical dataconcerning one or more time series.

    A time series is a set of observations measured at successivepoints in time or over successive periods of time.

    If the historical data used are restricted to past values of the seriesthat we are trying to forecast, the procedure is called a time seriesmethod.

    If the historical data used involve other time series that are believedto be related to the time series that we are trying to forecast, the

    procedure is called a causal method.Quantitative approaches are generally preferred. In this chapter wewill focus on quantitative approaches to forecasting.

  • 8/10/2019 New Forecasting Techniques

    7/46

  • 8/10/2019 New Forecasting Techniques

    8/46

    Dr. C. LightnerFayetteville State University

    8

    Components of a Time Series

    The trend component accounts for the gradual shiftingof the time series over a long period of time.

    Any regular pattern of sequences of values above andbelow the trend line is attributable to the cyclical

    component of the series.The seasonal component of the series accounts forregular patterns of variability within certain time periods,such as over a year.

    The irregular component of the series is caused byshort-term, unanticipated and non-recurring factors thataffect the values of the time series. One cannot attemptto predict its impact on the time series in advance.

  • 8/10/2019 New Forecasting Techniques

    9/46

    Dr. C. LightnerFayetteville State University

    9

    Time Series Data

    We will learn the following Forecasting Approaches:Smoothing

    Trend Projections

  • 8/10/2019 New Forecasting Techniques

    10/46

    Dr. C. LightnerFayetteville State University

    10

    Excel Instructions for Drawing a Scatter Plot

    1. Enter data in the Excel spreadsheet.2. Click on Inserton the toolbar and then click on the Chart tab. The

    Chart Wizard will appear. In step 1 on select the XY (scatter) chart

    type and then click next.

    3. In step 2 specify the cells where your data is located in the datarange box.

    4. In step 3 you can give your chart a title and label your axes. In

    step 4 specify where you want the chart to be placed.

  • 8/10/2019 New Forecasting Techniques

    11/46

    Dr. C. LightnerFayetteville State University

    11

    During the past ten weeks, sales of cases of Comfort brandheadache medicine at Robert's Drugs have been as follows:

    Week Sales Week Sales

    1 110 6 120

    2 115 7 1303 125 8 115

    4 120 9 110

    5 125 10 130

    Plot this data.

    Example: Roberts Drugs

  • 8/10/2019 New Forecasting Techniques

    12/46

    Dr. C. LightnerFayetteville State University

    12

    Plot Roberts Drugs Example

    Excel Spreadsheet Showing Input Data. Specify cells A4:B13 as the DataRange.

    A B

    1 Robert's Drugs

    2

    3 Week (t) Salest

    4 1 110

    5 2 115

    6 3 125

    7 4 120

    8 5 125

    9 6 120

    10 7 130

    11 8 115

    12 9 110

    13 10 130

    14 11

  • 8/10/2019 New Forecasting Techniques

    13/46

    Dr. C. LightnerFayetteville State University

    13

    Plot Roberts Drugs Example

    Robert's Drug Example

    105

    110

    115

    120

    125

    130

    135

    0 5 10 15

    Week, t

    Sales

    I labeledRoberts Drug

    Example as

    The Chart title

    I labeledWeek, t as

    My Value (x)

    axis

    I labeled

    Sales asMy Value (y)

    axis

  • 8/10/2019 New Forecasting Techniques

    14/46

    Dr. C. LightnerFayetteville State University

    14

    Smoothing Methods

    In cases in which the time series is fairly stable andhas no significant trend, seasonal, or cyclical effects,

    one can use smoothing methods to average out the

    irregular components of the time series.

    Three common smoothing methods are: Moving average

    Weighted moving average

    Exponential smoothing

  • 8/10/2019 New Forecasting Techniques

    15/46

    Dr. C. LightnerFayetteville State University

    15

    Smoothing Methods: Moving Average

    Moving Average MethodThe moving average method consists of computing

    an average of the most recent ndata values for the

    series and using this average for forecasting the value

    of the time series for the next period.

  • 8/10/2019 New Forecasting Techniques

    16/46

    Dr. C. LightnerFayetteville State University

    16

    Robert Drugs Example: Moving Average

    Our scatter plot for Roberts Drug Sales has nosignificant trend, seasonal, or cyclical effects. Thus we

    should employ a smoothing technique for forecasting

    sales.

    Forecast the sales for period 11 using a three period

    moving average (MA3).

  • 8/10/2019 New Forecasting Techniques

    17/46

    Dr. C. LightnerFayetteville State University

    17

    Example: Roberts Drugs: Moving Average

    Steps to Moving Average Using Excel

    Step 1: Select the Toolspull-down menu.

    Step 2: Select the Data Analysisoption.

    Step 3: When the Data Analysis Tools dialog

    appears, choose Moving Average.Step 4: When the Moving Average dialog box

    appears:

    Enter B4:B13 in the Input Rangebox.

    Enter 3 in the Intervalbox.Enter C5 in the Output Rangebox.

    Select OK.

    This specifies

    the value of n

    This is the column

    following our data,

    and one row below where

    our data begins.

  • 8/10/2019 New Forecasting Techniques

    18/46

    Dr. C. LightnerFayetteville State University

    18

    Roberts Drugs: Moving Average

    MA3(Three period Moving average) for Roberts Drug Example

    Ft is the forecast for week t.

    F4 (forecast for week 4)=116.7

    F11 (forecast for week 11)=118.3

    Thus we would forecast the sales

    for Week 11 to be 118.3

    Robert's Drugn=3

    Week (t ) Yt Ft

    1 110

    2 115 #N/A

    3 125 #N/A

    4 120 116.6667

    5 125 120

    6 120 123.3333

    7 130 121.6667

    8 115 1259 110 121.6667

    10 130 118.3333

    11 118.3333

  • 8/10/2019 New Forecasting Techniques

    19/46

    Dr. C. LightnerFayetteville State University

    19

    Smoothing Methods: Weighted Moving Average

    Weighted Moving Average MethodThe weighted moving average method consists of computing a

    weighted average of the most recent ndata values for the series

    and using this weighted average for forecasting the value of the

    time series for the next period. The more recent observations are

    typically given more weight than older observations. Forconvenience, the weights usually sum to 1.

    The regular moving average gives equal weight to past data values

    when computing a forecast for the next period. The weighted

    moving average allows different weights to be allocated to past

    data values.There is no Excel command for computing this so you must do this

    manually. You can either manually enter the formulas into excel

    and apply to all periods or compute value by hand.

  • 8/10/2019 New Forecasting Techniques

    20/46

    Dr. C. LightnerFayetteville State University

    20

    Smoothing Methods: Weighted Moving Average

    Use a 3 period weighted moving average to forecast the sales forweek 11 giving a weight of 0.6 to the most recent period, 0.3 to the

    second most recent period, and 0.1 to the third most recent period.

    F11= (0.6)*130 + (0.3)*110 + (0.1)* 115= 122.5

    Thus we would forecast the sales for week 11 to be 122.5.

    Sales for the

    most recent

    period

    Sales for 2nd

    most recent

    period

    Sales for 3rd

    most recent

    period

  • 8/10/2019 New Forecasting Techniques

    21/46

    Dr. C. LightnerFayetteville State University

    21

    Smoothing Methods: Exponential Smoothing

    Exponential Smoothing Using exponential smoothing, the forecast for the next period

    is equal to the forecast for the current period plus a

    proportion () of the forecast error in the current period.

    Using exponential smoothing, the forecast is calculated by:

    Ft+1= Yt + (1- )Ft

    where:

    is the smoothing constant (a number between 0 and 1)

    Ft is the forecast for period tFt +1 is the forecast for period t+1

    Yt is the actual data value for period t

    This is the same as

    Ft+1 = Ft+ (YtFt)

  • 8/10/2019 New Forecasting Techniques

    22/46

    Dr. C. LightnerFayetteville State University

    22

    Roberts Drugs: Exponential Smoothing

    Forecast the sales for period 11 using ExponentialSmoothing = 0.1.

  • 8/10/2019 New Forecasting Techniques

    23/46

    Dr. C. LightnerFayetteville State University 23

    Roberts Drugs: Exponential Smoothing

    Steps to Exponential Smoothing Using Excel

    Step 1: Select the Toolspull-down menu.

    Step 2: Select the Data Analysisoption.

    Step 3: When the Data Analysis Tools dialog

    appears, choose Exponential Smoothing.Step 4: When the Exponential Smoothing dialog box

    appears:

    Enter B4:B13 in the Input Rangebox.

    Enter 0.9 (for = 0.1) in Damping Factorbox.

    Enter C4 in the Output Rangebox.

    Select OK.

    Damping factor

    is always 1-

  • 8/10/2019 New Forecasting Techniques

    24/46

    Dr. C. LightnerFayetteville State University 24

    Roberts Drugs: Exponential Smoothing

    F11 = 0.1 * Y10 + .9 F10= .1 *130 + .9 * 115.4099

    = 116.87

    Robert's Drugs

    =0.1

    Week (t) Salest Ft

    1 110 #N/A

    2 115 110

    3 125 110.5

    4 120 111.95

    5 125 112.755

    6 120 113.9795

    7 130 114.5816

    8 115 116.1234

    9 110 116.0111

    10 130 115.4099

    11

    Thus we would

    forecast sales for

    week 11 to be 116.87

  • 8/10/2019 New Forecasting Techniques

    25/46

    Dr. C. LightnerFayetteville State University 25

    Questions That You Should Be Asking

    For the Moving Average technique, how do I determine the bestvalue of n to use for forecasting?

    For Exponential Smoothing, how do I determine the best value of

    to use?

    If I realize that a smoothing technique should be employed, how do

    you know which smoothing technique is best?

    In order to answer the above questions, we need criteria for

    judging the accuracy of a forecasting technique. Once we select a

    criterion, the method (or parameter) which provides the best value

    for our criterion is the best method (or parameter) to use forforecasting our scenario.

  • 8/10/2019 New Forecasting Techniques

    26/46

    Dr. C. LightnerFayetteville State University 26

    Measures of Forecast Accuracy

    Mean Squared Error (MSE)The average of the squared forecast errors for the historical

    data is calculated. The forecasting method or parameter(s) which

    minimize this mean squared error is then selected.

    Mean Absolute Deviation (MAD)The mean of the absolute values of all forecast errors is

    calculated, and the forecasting method or parameter(s) which

    minimize this measure is selected. The mean absolute deviation

    measure is less sensitive to individual large forecast errors than the

    mean squared error measure.

    You may choose either of the above criteria for evaluating the

    accuracy of a method (or parameter).

  • 8/10/2019 New Forecasting Techniques

    27/46

    Dr. C. LightnerFayetteville State University 27

    Selecting the best Smoothing Technique for Roberts Drugs

    Determine the smoothing technique that is best for forecastingRoberts Drug sales: A two period moving average, a three period

    moving average, exponential smoothing (=0.1), or exponential

    smoothing (=0.2)

    Realistically we should have experimented with more values of n

    for the moving average, and for exponential smoothing to

    determine the absolute best parameters to use for our technique.

    On the next slide we randomly chose to use the MSE criterion tojudge the best technique.

  • 8/10/2019 New Forecasting Techniques

    28/46

  • 8/10/2019 New Forecasting Techniques

    29/46

    Dr. C. LightnerFayetteville State University 29

    Roberts Drugs :Comparing Smoothing Techniques

    MSE for MA3

    Robert's Drug

    Sales n=3 Error

    Week (t ) Yt Ft (Yt- Ft) (Yt- Ft)2

    1 110

    2 115 #N/A

    3 125 #N/A

    4 120 116.6667 3.333333 11.111115 125 120 5 25

    6 120 123.3333 -3.33333 11.11111

    7 130 121.6667 8.333333 69.44444

    8 115 125 -10 100

    9 110 121.6667 -11.6667 136.1111

    10 130 118.3333 11.66667 136.1111

    11 118.3333

    MSE 69.84127

  • 8/10/2019 New Forecasting Techniques

    30/46

    Dr. C. LightnerFayetteville State University 30

    Roberts Drugs :Comparing Smoothing Techniques

    MSE for Exponential

    Smoothing =0.1

    Sales =0.1 Error

    Week (t ) Yt Ft (Yt- Ft) (Yt- Ft)2

    1 110 #N/A

    2 115 110 5 25

    3 125 110.5 14.5 210.25

    4 120 111.95 8.05 64.8025

    5 125 112.755 12.245 149.94

    6 120 113.9795 6.0205 36.24642

    7 130 114.5816 15.41845 237.7286

    8 115 116.1234 -1.1234 1.262016

    9 110 116.0111 -6.01106 36.13279

    10 130 115.4099 14.59005 212.8696

    11

    MSE 108.248

  • 8/10/2019 New Forecasting Techniques

    31/46

    Dr. C. LightnerFayetteville State University 31

    Roberts Drugs :Comparing Smoothing Techniques

    MSE for Exponential

    Smoothing =0.2

    Sales =0.2 Error

    Week (t ) Yt Ft (Yt- Ft) (Yt- Ft)2

    1 110 #N/A

    2 115 110 5 25

    3 125 111 14 196

    4 120 113.8 6.2 38.44

    5 125 115.04 9.96 99.2016

    6 120 117.032 2.968 8.809024

    7 130 117.6256 12.3744 153.1258

    8 115 120.1005 -5.10048 26.0149

    9 110 119.0804 -9.08038 82.45337

    10 130 117.2643 12.73569 162.1979

    11

    MSE 87.91584

  • 8/10/2019 New Forecasting Techniques

    32/46

    Dr. C. LightnerFayetteville State University 32

    Roberts Drugs :Comparing Smoothing Techniques

    Since the three period moving average technique(MA3) provides to lowest MSE value, this is the best

    smoothing technique to use for forecasting Roberts

    Drug Sales.

  • 8/10/2019 New Forecasting Techniques

    33/46

    Dr. C. LightnerFayetteville State University 33

    Trend Projection

    If a time series exhibits a linear trend, the method of leastsquares may be used to determine a trend line (projection) for

    future forecasts.

    Least squares, also used in regression analysis, determines the

    unique trend line forecast which minimizes the mean square

    error between the trend line forecasts and the actual observedvalues for the time series.

    The independent variable is the time period and the dependent

    variable is the actual observed value in the time series.

  • 8/10/2019 New Forecasting Techniques

    34/46

    Dr. C. LightnerFayetteville State University 34

    Trend Projection

    Using the method of least squares, the formula for the trendprojection is:Y

    t= b0+ b1t.

    where: Yt= trend forecast for time period t

    b1 = slope of the trend line

    b0= trend line projection for time 0

    b1= n tYt- tYt

    nt 2- (t )2

    where: Yt= observed value of the time series at time period t

    = average of the observed values for Yt

    = average time period for the n observations

    0 1b Y b t

    t

    t Y

    t

  • 8/10/2019 New Forecasting Techniques

    35/46

    Dr. C. LightnerFayetteville State University 35

    Example: Augers Plumbing Service

    The number of plumbing repair jobs performed by Auger's Plumbing

    Service in each of the last nine months are listed below.

    Month Jobs Month Jobs Month Jobs

    March 353 June 374 September 399

    April 387 July 396 October 412May 342 August 409 November 408

    Forecast the number of repair jobs Auger's will perform in

    December using the least squares method.

  • 8/10/2019 New Forecasting Techniques

    36/46

    Dr. C. LightnerFayetteville State University 36

    Augers Plumbing Service: Trend Projection

    Trend Projection

    (month) t Yt tYt t2

    (Mar.) 1 353 353 1

    (Apr.) 2 387 774 4

    (May) 3 342 1026 9(June) 4 374 1496 16

    (July) 5 396 1980 25

    (Aug.) 6 409 2454 36

    (Sep.) 7 399 2793 49(Oct.) 8 412 3296 64

    (Nov.) 9 408 3672 81

    Sum 45 3480 17844 285

  • 8/10/2019 New Forecasting Techniques

    37/46

    Dr. C. LightnerFayetteville State University 37

    Example: Augers Plumbing Service

    Trend Projection (continued)

    = 45/9 = 5 = 3480/9 = 386.667

    ntYt- t Yt (9)(17844) - (45)(3480)b

    1= = = 7.4n t 2- (t)2 (9)(285) - (45)2

    = 386.667 - 7.4(5) = 349.667

    Thus our trend line is Yt= 349.667+ 7.4 t.

    Y10= 349.667 + (7.4)(10) = 423.667

    0 1b Y b t

    Yt t

    For December t=10

  • 8/10/2019 New Forecasting Techniques

    38/46

    Dr. C. LightnerFayetteville State University 38

    Augers Plumbing Service: Trend Line in Excel

    A B C

    1 Auger's Plumbing Service

    2

    3 Month Calls

    4 1 3535 2 387

    6 3 342

    7 4 374

    8 5 396

    9 6 40910 7 399

    11 8 412

    12 9 408

    13

    Excel Spreadsheet Showing Input Data

  • 8/10/2019 New Forecasting Techniques

    39/46

    Dr. C. LightnerFayetteville State University 39

    Example: Augers Plumbing Service

    Steps to Trend Projection Using ExcelStep 1: Select an empty cell (B13) in the worksheet.

    Step 2: Select the Insertpull-down menu.

    Step 3: Choose the Functionoption.

    Step 4: When the Select Category dialog box appears:

    Choose Statisticalin Function Category box.

    Choose Forecastin the Function Name box.

    Select OK.

    Step 5: When the Forecast dialog box appears:

    Enter 10 in the xbox (for month 10).Enter B4:B12 in the Known ysbox.

    Enter A4:A12 in the Known xsbox.

    Select OK.

  • 8/10/2019 New Forecasting Techniques

    40/46

    Dr. C. LightnerFayetteville State University 40

    Example: Augers Plumbing Service

    Spreadsheet Showing Trend Projection for Month 10Auger's Plumbing Service

    Month Calls

    1 353

    2 3873 342

    4 374

    5 396

    6 409

    7 3998 412

    9 408

    10 423.667 Projected

  • 8/10/2019 New Forecasting Techniques

    41/46

    Dr. C. LightnerFayetteville State University 41

    Roberts Drug Example

    Suppose we neglected to plot Roberts Drug example, and therefore wedo not know that a trend does not exist. Use trend analysis to forecast

    the sales for month 11.

    Week (t) Yt1 110

    2 115

    3 125

    4 120

    5 125

    6 120

    7 130

    8 115

    9 110

    10 130

    11 124 Forecast

  • 8/10/2019 New Forecasting Techniques

    42/46

    Dr. C. Lightner

    Fayetteville State University

    42

    Question????

    How could we use the MSE or MAD to verify that theMA3is a better smoothing technique than trend analysis

    for Roberts Drug Sales data?

  • 8/10/2019 New Forecasting Techniques

    43/46

    Dr. C. Lightner

    Fayetteville State University

    43

    Causal Method: Regression Analysis

    Regression Analysis is similar to trend analysis, exceptthe independent variable is not restricted to time. Refer

    to Roberts Drug example. Instead of letting time

    represent our independent variable, we could forecast

    sales based upon the price of the product. Since

    products often go on sale, we could collect data over

    several months collecting the weekly price and number

    of items sold for the week. For this model, we would

    find the regression equation in the same manner in

    which we found the trend line except we would call theindependent variable x, instead of t.

  • 8/10/2019 New Forecasting Techniques

    44/46

    Dr. C. Lightner

    Fayetteville State University

    44

    Regression Equation

    Using the method of least squares, the formula for the regressionline is:Y = b0+ b1x.

    where: Y= dependent variable which depends on the value of x

    b1 = slope of the regression line

    b0= regression line projection for x= 0

    b1= n XiYi- XiYi

    nXi2- (Xi)

    2

    where: Yt= observed value of the time series at time period t

    = average of the observed values for Yt

    = average time period for the n observations

    t

    t

    Y

    t

    b y b x0 1

  • 8/10/2019 New Forecasting Techniques

    45/46

    Dr. C. Lightner

    Fayetteville State University

    45

    Regression Analysis in Excel

    The dependent variable Y can predicted using thesame forecast function in Excel as used to forecast a

    trend line. Follow the same steps provided on slide 39.

  • 8/10/2019 New Forecasting Techniques

    46/46

    Dr. C. Lightner 46

    THE END

    See your textbook for moreexamples and detailed explanations

    of all topics discussed in these notes.