Lecture.chap15

download Lecture.chap15

of 32

Transcript of Lecture.chap15

  • 8/3/2019 Lecture.chap15

    1/32

    Chapter 15 Ex panded Notes

    Slide 1

    1 2003 Thomson/South-Western Slide

    Chapter 15Forecasting

    Qualitative 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

    We will cover the above-mentioned forecasting methods, covering sections 1 through 3 andsection 6. Read over section 5 (Regression analysis which you must have covered in statisticsclass)

    Slide 2

    Defining forecasting

    2 2003 Thomson/South-Western Slide Heizer Render 2001 by Prentice

    Hall, Inc.,

    4-2

    What is Forecasting?

    Process of predicting a

    future event

    Underlying basis of

    all business decisions Production

    Inventory

    Personnel

    Facilities

    Sales will be$200 Million!

  • 8/3/2019 Lecture.chap15

    2/32

    Slide 3

    3 2003 Thomson/South-Western Slide Heizer Render,2001 by

    Prentice Hall,

    4-3

    Short-range forecast

    Up to 1 year; usually less than 3 months

    Job scheduling, worker assignments

    Medium-range forecast

    3 months to 3 years

    Sales & production planning, budgeting

    Long-range forecast

    3+ years

    New product planning, facility location

    Types of Forecasts by Time Horizon

    Slide 4

    4 2003 Thomson/South-Western Slide Heizer Render, 2001 by Prentice

    Hall, Inc.,

    4-4

    Forecasting Approaches

    Used when situation is

    stable & historical data

    exist Existing products

    Current technology

    Involves mathematical

    techniques

    e.g., forecasting sales of

    color televisions

    Quantitative Methods Used when situation is

    vague & littl e data exist New product s

    New technol ogy

    Involves intuition,

    experience

    e.g., forecasting sales on

    Internet

    Qualitative Methods

    Forecasting methods fall under two main umbrellas: quantitative based and qualitative based.

    The slide provides the main difference between the two approaches and their applicability.

  • 8/3/2019 Lecture.chap15

    3/32

    Slide 5

    5 2003 Thomson/South-Western Slide

    Qualitative Approaches to Forecasting

    Delphi Approach

    A panel of experts, each of whom is physicallyseparated from the others and is anonymous, isasked to respond to a sequential series ofquestionnaires.

    After each questionnaire, the responses are tabulatedand the information and opinions of the entire groupare made known to each of the other panel membersso that they may revise their previous forecastresponse.

    The process continues until some degree ofconsensus is achieved.

    Qualitative approaches to forecasting can take on a variety of different methods. In the DelphiApproach a panel of experts is used to determine what the forecast will be. A good example isthe forecasted price of stocks. Stock price forecasts are based on expert opinions from variousfund managers and industry experts.

    Slide 6

    6 2003 Thomson/South-Western Slide

    Qualitative Approaches to Forecasting

    Scenario WritingScenario writing consists of developing a conceptual

    scenario of the future based on a well defined set ofassumptions.

    After several different scenarios have beendeveloped, the decision maker determines which ismost likely to occur in the future and makesdecisions accordingly.

    Scenario writing is another approach to qualitative forecasting. Here a variety of scenarios arewritten and the decision maker determines what he / she feels is most likely. A businessmanager might use this to gather input from each of the different business areas. For example,production might write a scenario, sales might write another scenario, and finance might write ascenario. The decision maker would then evaluate each of the potential scenarios anddetermine the forecast.

  • 8/3/2019 Lecture.chap15

    4/32

    Slide 7

    7 2003 Thomson/South-Western Slide

    Qualitative Approaches to Forecasting

    Subjective or Interactive Approaches

    These techniques are often used by committees orpanels seeking to develop new ideas or solvecomplex problems.

    They often involve "brainstorming sessions".

    It is important in such sessions that any ideas oropinions be permitted to be presented withoutregard to its relevancy and without fear of criticism.

    Subjective approaches typically involve brainstorming to determine what the committee or groupviews as the likely forecast. A great example is when the sales manager asks each employee toestimate their likely sales. The group then discusses their potential as a whole and from thisdata the sales manager will forecast his sales.

    Slide 8

    8 2003 Thomson/South-Western Slide

    Quantitative Approaches to Forecasting

    Quantitative methods are based on an analysis ofhistorical data concerning one or more time series.

    A time series is a set of observations measured atsuccessive points in time or over successive periodsof time.

    If the historical data used are restricted to past valuesof the series that we are trying to forecast, theprocedure is called a time series method.

    If the historical data used involve other time seriesthat are believed to be related to the time series thatwe are trying to forecast, the procedure is called acausal method.

    Although qualitative methods have their purpose, quantitative methods are rooted in data andshould provide a better forecast. It is important to use good, historical data that depicts what hashappened and can likely predict what will happen. We will study a number of time seriesmethods for quantitative forecasting and we will use regression analysis to develop a causalforecast.

  • 8/3/2019 Lecture.chap15

    5/32

    Slide 9

    9 2003 Thomson/South-Western Slide

    Quantitative Forecasting Methods

    Quantitative

    Forecasting

    Linear

    Regression

    Associative

    Models

    ExponentialSmoothing

    MovingAverage

    Time Series

    Models

    TrendProjection

    Heizer Render, 2001 by Prentice Hall,

    Inc

    Quantitative methods in forecasting fall in turn into 2 main streams: time series based and

    Associative or Causal Model. Linear regression is the most popular causal method used (read

    over section 5). We will focus on times series models.

  • 8/3/2019 Lecture.chap15

    6/32

    Slide 10

    10 2003 Thomson/South-Western Slide

    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 cyclicalcomponent of the series.

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

    The irregular component of the series is caused byshort-term, unanticipated and non-recurring factorsthat affect the values of the time series. One cannotattempt to predict its impact on the time series in

    advance.

    Time series implies that the historical data have been gathered over a defined period of regulartime intervals. This could be weeks, months, days, etc. Plotting the data is always the best step.A simple line chart will show if there is any pattern to the data. Patterns might be the result ofseasonal increases in sales or regular, cyclical occurring influencers. I always think about myhusbands company who purchases a significant number of cars for their employees. The carpurchasing is cyclical every three years new cars are bought. If you are the lucky dealershipfrom whom these cars are purchased you would show a cyclical (every 3 years) pattern for thiscustomer. Car dealerships might also show seasonal patterns, for example near the end of the

    year when they have significant sales in an effort to reduce their inventory (they do this for taxpurposes, so I am told ~ not really because of the volume of incoming new models!)

  • 8/3/2019 Lecture.chap15

    7/32

    Slide 11

    11 2003 Thomson/South-Western Slide Heizer Render, 2001 by Prentice

    Hall, Inc.,

    4-11

    Trend

    Seasonal

    Cyclical

    Random

    Time Series Components

    Slide summarizes again the main time series components

  • 8/3/2019 Lecture.chap15

    8/32

    Slide 12

    12 2003 Thomson/South-Western Slide4-12

    Persistent, overall upward or downward pattern

    Due to population, technology etc.

    Several years duration

    Mo., Qtr., Yr.

    Response

    1984-1994T/Maker Co.

    Trend Component

    Heizer Render, 2001 by Prentice

    Hall, Inc.,

    Slide 13

    13 2003 Thomson/South-Western Slide4-13

    Regular pattern of up & down fluctuations Due to weather, customs etc.

    Occurs within 1 year

    Mo., Qtr.

    Response

    Summer

    1984-1994T/Maker Co.

    Seasonal Component

    Heizer Render, 2001 by Prentice

    Hall, Inc.,

  • 8/3/2019 Lecture.chap15

    9/32

    Slide 14

    14 2003 Thomson/South-Western Slide4-14

    Repeating up & down movements

    Due to interactions of factors influencing economy

    Usually 2-10 years duration

    Mo., Qtr., Yr.

    Response

    Cycle

    Cyclical Component

    Heizer Render, 2001 by Prentice

    Hall, Inc.,

    Slide 15

    15 2003 Thomson/South-Western Slide4-15

    Erratic, unsystematic, residual fluctuations Due to random variation or unforeseen events

    Union strike

    Tornado

    Short duration &nonrepeating

    1984-1994 T/Maker Co.

    Random Component

    Heizer Render, 2001 by Prentice

    Hall, Inc.,

  • 8/3/2019 Lecture.chap15

    10/32

    Slide 16

    16 2003 Thomson/South-Western Slide

    Time Series Methods

    Three time series methods are:

    smoothing

    trend projection

    trend projection adjusted for seasonal influence

    These are the three time series methods that we will review.

  • 8/3/2019 Lecture.chap15

    11/32

    Slide 17

    17 2003 Thomson/South-Western Slide

    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 theirregular components of the time series.

    Three common smoothing methods are:

    Moving averages

    Weighted moving averages

    Exponential smoothing

    Our goal in forecasting is to smooth out the irregular patterns and plan for the known seasonalor cyclical patterns. Three common smoothing methods are: Moving averages, weighted movingaverages, and exponential smoothing. Each of these methods has a different approach tosmoothing out the patterns in the data.

  • 8/3/2019 Lecture.chap15

    12/32

    Slide 18

    18 2003 Thomson/South-Western Slide

    Smoothing Methods

    Moving Average Method

    The moving average method consists of computingan average of the most recent n data values for theseries and using this average for forecasting the value ofthe time series for the next period.

    The moving average method is the least complex. Basically you use the average of some periodof recent data to predict future data. Lets take a look through an example.

    Slide 19

    19 2003 Thomson

    /South-Western Slide

    During the past ten weeks, sales of cases ofComfort brand headache medicine at Robert's Drugshave been as follows:

    Week Sales Week Sales1 110 6 1202 115 7 1303 125 8 1154 120 9 1105 125 10 130

    If Robert's uses a 3-period moving average toforecast sales, what is the forecast for Week 11?

    Example: Roberts Drugs

    Roberts Drugs shows their sales data for the last 10 weeks. They would like to use a 3-periodmoving average to forecast week 11. To do this, the forecaster will use the historical data forweeks 8, 9, and 10 and determine the average. This value is the forecast for week 11.

  • 8/3/2019 Lecture.chap15

    13/32

    Slide 20

    20 2003 Thomson/South-Western Slide

    Example: Rosco Drugs

    Excel Spreadsheet Showing Input Data

    A B C

    1 Robert's Drugs

    2

    3 Week (t) Salest Forect+14 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

    Slide 21

    21 2003 Thomson/South-Western Slide

    Example: Rosco Drugs

    Steps to Moving Average Using Excel

    Step 1: Select the Tools pull-down menu.Step 2: Select the Data Analysis option.

    Step 3: When the Data Analysis Tools dialogappears, choose Moving Average.

    Step 4: When the Moving Average dialog boxappears:

    Enter B4:B13 in the Input Range box.

    Enter 3 in the Interval box.

    Enter C4 in the Output Range box.

    Select OK.

    Excel has a function that easily computes the moving average. See instructions in slide above.For Excell 2007, need to upload the ToolPack Analysis tool. Will show in class and will uploadinstructions on shell.

  • 8/3/2019 Lecture.chap15

    14/32

    Slide 22

    22 2003 Thomson/South-Western Slide

    Example: Roberts Drugs

    Spreadsheet Showing Results Using n = 3

    A B C

    1 Robert's Drugs

    2

    3 Week (t) Salest Forect+14 1 110 #N/A

    5 2 115 #N/A

    6 3 125 116.7

    7 4 120 120.0

    8 5 125 123.3

    9 6 120 121.7

    10 7 130 125.0

    11 8 115 121.7

    12 9 110 118.313 10 130 118.3

    Excel has a function that easily computes the moving average. In the above slide you can see

    that excel has computed the forecasted 3-period moving average for weeks 4-14. It is helpful to

    have a forecast for historical data ~ that is, we did not need to forecast week 4, because we

    already knew the sales figures (Remember from the previous slide we had data through week

    10). But by doing this, we can determine the accuracy of our forecast methodology. We will look

    at measures of forecast accuracy later in this lecture.

    Take a minute and notice how Excel outputs the data. The forecast that is horizontal with week

    3 is 116.7. This is actually the forecast for week 4 so, if we wanted to see how well our model

    would have predicted the forecast for week 4, we can compare the forecast values for week 4

    (116.7) with the actual values for week 4 (120).

  • 8/3/2019 Lecture.chap15

    15/32

    Slide 23

    23 2003 Thomson/South-Western Slide

    Smoothing Methods

    Weighted Moving Average Method

    In the weighted moving average method forcomputing the average of the most recent n periods,the more recent observations are typically given moreweight than older observations. For convenience, theweights usually sum to 1.

    The weighted moving average is similar to the moving average except we assign weights toeach of the time periods. Remember our goal: to smooth out variations in the data that areirregular. We might find that our sales show an increasing pattern. In this case, using a movingaverage of 3-periods will not account for the increasing pattern. Instead, we might assign a highweight to the most recent period and a lower weight to the less recent periods. Lets look atan example.

  • 8/3/2019 Lecture.chap15

    16/32

    Slide 24

    24 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service

    Forecast for December (Month 10) using a three-period (n = 3) weighted moving average with weightsof .6, .3, and .1.

    Month Jobs

    March 353

    April 387

    May 342

    June 374

    July 396

    Aug 409

    Sept 399

    Oct 412

    Nov 408

    Lets use a 3-period weighted moving average to forecast Augers plumbing service. Theweights are given above.

    Slide 25

    25 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service (B)

    Three-Month Weighted Moving Average

    The forecast for December will be the weightedaverage of the preceding three months: September,

    October, and November.F10 = .1YSep. + .3YOct. + .6YNov.

    = .1(399) + .3(412) + .6(408)

    = 408.3

    The forecast for period 10 (December) uses the previous 3 months of data (Sept, Oct, Nov).November is the most recent month so it is assigned a weight of .6. October is assigned aweight of .3 and September is assigned a weight of .1. We will multiple each of the previousmonths actual data times the weight that has been assigned to that month. The resulting value,408.3, is the forecast for December.

    Although this is not difficult to do in Excel, you must manually set up the equations. Excel doesnot have a function that computes the weighted moving average.

  • 8/3/2019 Lecture.chap15

    17/32

    Slide 26

    26 2003 Thomson/South-Western Slide

    Smoothing Methods

    Exponential Smoothing

    Using exponential smoothing, the forecast for thenext period is equal to the forecast for the currentperiod plus a proportion () of the forecast errorin the current period.

    Using exponential smoothing, the forecast iscalculated by:

    [the actual value for the current period] +

    (1- )[the forecasted value for the current period],

    where the smoothing constant, , is a numberbetween 0 and 1. OR

    F = Y + Ft t-1 (1- )t-1

    The third smoothing method that we will consider is exponential smoothing. This method is veryuseful for companies with little historical data. Exponential smoothing only requires one monthof historical data. The formula is shown above in a yellow box. The forecast for period t is equalto the value of the smoothing constant times the actual sales for the previous month plus 1minus the smoothing constant times the forecast for the previous month.

    Exponential smoothing uses a smoothing constant and the accuracy of the previous monthsforecast to determine the forecast for the next period. The smoothing constant is determined by

    the decision maker. It is a value between 0 and 1. The closer the value is to 1 the more theemphasis is on the previous months forecast (and more recent periods). Lets look at anexample.

  • 8/3/2019 Lecture.chap15

    18/32

    Slide 27

    27 2003 Thomson/South-Western Slide

    Example: Roberts Drugs

    During the past ten weeks, sales of cases ofComfort brand headache medicine at Robert's Drugshave been as follows:

    Week Sales Week Sales1 110 6 1202 115 7 1303 125 8 1154 120 9 1105 125 10 130

    If Robert's uses exponential smoothing toforecast sales, which value for the smoothingconstant , = .1 or = .8, gives better forecasts?

    Roberts Drugs is not convinced that his moving average forecast is good. He would like to useexponential smoothing to forecast his sales.Robert is going to use an exponential smoothing constant of .1 and .8. He plans to compare theaccuracy of each forecast and determine the best one.

  • 8/3/2019 Lecture.chap15

    19/32

    Slide 28

    28 2003 Thomson/South-Western Slide

    Example: Roberts Drugs

    Exponential Smoothing ( = .1, 1 - = .9)

    F1 = 110F2 = .1Y1 + .9F1 = .1(110) + .9(110) = 110

    F3 = .1Y2 + .9F2 = .1(115) + .9(110) = 110.5

    F4 = .1Y3 + .9F3 = .1(125) + .9(110.5) = 111.95

    F5 = .1Y4 + .9F4 = .1(120) + .9(111.95) = 112.76

    F6 = .1Y5 + .9F5 = .1(125) + .9(112.76) = 113.98

    F7 = .1Y6 + .9F6 = .1(120) + .9(113.98) = 114.58

    F8 = .1Y7 + .9F7 = .1(130) + .9(114.58) = 116.12

    F9 = .1Y8 + .9F8 = .1(115) + .9(116.12) = 116.01

    F10= .1Y9 + .9F9 = .1(110) + .9(116.01) = 115.41

    The forecast for period 1 is 110. This is because the forecast for the first period is always setequal to the actual for that period (unless otherwise given in the problem). This does force theforecast for period 2 to be equal to the first months actual data, but after the first 2 periods theforecast works fine. So to forecast period 2, Robert multiplies the smoothing constant of .1 timesthe actual sales for period 1 (110). He adds to this the multiplication of .9 (1 - .1) to the forecastfor period 1. This process continues Lets focus in on period 10s forecast. The forecast forperiod 10 is .1 times the actual for period 9 (110) plus .9 times the forecast for period 9 (116.01).

    If you look closely you can see that a smoothing constant close to 1 will put a lot of emphasis onthe actual value for the previous month. So the smoothing constant that you choose isdependent upon how much you want to rely on the previous months data versus the previouslyforecasted data.

  • 8/3/2019 Lecture.chap15

    20/32

    Slide 29

    29 2003 Thomson/South-Western Slide

    Example: Roberts Drugs

    Exponential Smoothing ( = .8, 1 - = .2)

    F1 = 110

    F2 = .8(110) + .2(110) = 110

    F3 = .8(115) + .2(110) = 114

    F4 = .8(125) + .2(114) = 122.80

    F5 = .8(120) + .2(122.80) = 120.56

    F6 = .8(125) + .2(120.56) = 124.11

    F7 = .8(120) + .2(124.11) = 120.82

    F8 = .8(130) + .2(120.82) = 128.16

    F9 = .8(115) + .2(128.16) = 117.63F10= .8(110) + .2(117.63) = 111.53

    Using an exponential smoothing constant of .8, we can determine another forecast for RobertsDrugs. In this example, the smoothing constant is close to 1, so a lot of emphasis is beingplaced on the actual sales data for the previous month.

    Exponential smoothing can be done easily in Excel. The next slides show how to set up in Exceland an excel poutput for the example.

  • 8/3/2019 Lecture.chap15

    21/32

    Slide 30

    30 2003 Thomson/South-Western Slide

    Example: Rosco Drugs (B)

    Excel Spreadsheet Showing Input Data

    A B C

    1 Robert's Drugs

    2

    3 Week Sales

    4 1 110

    5 2 115

    6 3 125

    7 4 120

    8 5 125

    9 6 120

    10 7 13011 8 115

    12 9 110

    13 10 130

    Slide 31

    31 2003 Thomson/South-Western Slide

    Example: Rosco Drugs (B)

    Steps to Exponential Smoothing Using Excel

    Step 1: Select the Tools pull-down menu.

    Step 2: Select the Data Analysis option.

    Step 3: When the Data Analysis Tools dialogappears, choose Exponential Smoothing.

    Step 4: When the Exponential Smoothing dialog boxappears:

    Enter B4:B13 in the Input Range box.

    Enter 0.9 (for = 0.1) in Damping Factor box.

    Enter C4 in the Output Range box.

    Select OK.

  • 8/3/2019 Lecture.chap15

    22/32

    Slide 32

    32 2003 Thomson/South-Western Slide

    Example: Rosco Drugs (B)

    Spreadsheet Showing Results Using = 0.1

    A B C

    1 Robert's Drugs

    2 = 0.1

    3 Week (t) Salest Forect+1

    4 1 110 #N/A

    5 2 115 110.0

    6 3 125 110.5

    7 4 120 112.0

    8 5 125 112.8

    9 6 120 114.0

    10 7 130 114.611 8 115 116.1

    12 9 110 116.0

    13 10 130 115.4

    Slide 33

    33 2003 Thomson/South-Western Slide

    Example: Rosco Drugs (B)

    Repeating the Process for = 0.8

    Step 4: When the Exponential Smoothing dialog boxappears:

    Enter B4:B13 in the Input Range box.

    Enter 0.2 (for = 0.8) in Damping Factor box.

    Enter D4 in the Output Range box.

    Select OK.

  • 8/3/2019 Lecture.chap15

    23/32

    Slide 34

    34 2003 Thomson/South-Western Slide

    Example: Rosco Drugs (B)

    Spreadsheet Results for = 0.1 and = 0.8

    A B C D

    1 Robert's Drugs

    2 = 0.1 = 0.8

    3 Week (t) Salest Forect+1 Forect+1

    4 1 110 #N/A #N/A

    5 2 115 110.0 110.0

    6 3 125 110.5 114.0

    7 4 120 112.0 122.8

    8 5 125 112.8 120.6

    9 6 120 114.0 124.1

    10 7 130 114.6 120.8

    11 8 115 116.1 128.2

    12 9 110 116.0 117.613 10 130 115.4 111.5

    Slide 35

    35 2003 Thomson/South-Western Slide Heizer Render, 2001 by Prentice Hall, Inc.4-35

    Seven Steps in Forecasting

    Select the items to be forecast Determine the time horizon of the forecast

    Collect Data

    Plot and analyze

    Select the forecasting model(s)

    Make the forecast

    Validate and implement results (monitoring)

    Accuracy (magnitude of error) and bias

    This slides puts forecasting process in perspective

  • 8/3/2019 Lecture.chap15

    24/32

    Slide 36

    36 2003 Thomson/South-Western Slide

    Measures of Forecast Accuracy

    Mean Squared Error

    The average of the squared forecast errors for thehistorical data is calculated. The forecasting method orparameter(s) which minimize this mean squared erroris then selected.

    Mean Absolute Deviation

    The mean of the absolute values of all forecasterrors is calculated, and the forecasting method orparameter(s) which minimize this measure is selected.The mean absolute deviation measure is less sensitive

    to individual large forecast errors than the meansquared error measure.

    Now that Robert has two different forecasts using exponential smoothing, we need to evaluatewhich forecast is more accurate. There are two approaches to forecast accuracy. If we subtractthe forecast value from the actual value we have an indication of how well the forecastperformed for each period. But lets get an overall measure of accuracy. One

    method is the

    mean square errors. Here we square the forecast errors for each period and then take theiraverage. The goal is to minimize your MSE. The issue with MSE is that your MSE is in squaredterms, so it is not always easy to interpret.

    One other approach is the mean absolute deviation. To compute the MAD, we take the absolutevalue of the forecast errors for each period and average these values. The MAD is easier tointerpret than the MAD because the units are not squared. The MAD provides an averageamount of deviation (over or under) that is given with the forecast. Again, the goal is to minimizethe MAD.

  • 8/3/2019 Lecture.chap15

    25/32

    Slide 37

    37 2003 Thomson/South-Western Slide

    Example: Roberts Drugs

    Mean Squared Error

    In order to determine which smoothing constantgives the better performance, calculate, for each, themean squared error for the nine weeks of forecasts,weeks 2 through 10 by:

    [(Y2-F2)2 + (Y3-F3)

    2 + (Y4-F4)2 + . . . + (Y10-F10)

    2]/9

    Using Roberts Drugs we can determine the MSE for both the smoothing constant of .1 and .8

    Slide 38

    38 2003 Thomson/South-Western Slide

    Example: Roberts Drugs

    = .1 = .8

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

    2

    1 1102 115 110.00 25.00 110.00 25.003 125 110.50 210.25 114.00 121.00

    4 120 111.95 64.80 122.80 7.845 125 112.76 149.94 120.56 19.716 120 113.98 36.25 124.11 16.917 130 114.58 237.73 120.82 84.238 115 116.12 1.26 128.16 173.309 110 116.01 36.12 117.63 58.26

    10 130 115.41 212.87 111.53 341.27

    Sum 974.22 Sum 847.52MSE Sum/9 Sum/9108.25 94.17

    The above table shows that the smoothing constant of .8 provides a lower MSE, thus it ispreferred to the smoothing constant of .1.

  • 8/3/2019 Lecture.chap15

    26/32

    Slide 39

    39 2003 Thomson/South-Western Slide

    Trend Projection

    If a time series exhibits a linear trend, the method ofleast squares 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 whichminimizes the mean square error between the trendline forecasts and the actual observed values for thetime series.

    The independent variable is the time period and thedependent variable is the actual observed value inthe time series.

    Trend projection is a time series method that uses the least squares to determine a trend line.Least squares method is also used in regression. Rather than have you labor throughcomputing a trend line by hand, I prefer to use regression to demonstrate both trend projectionand causal method forecasting.

    For regression, an independent variable is used to predict a dependent variable. In forecasting

    the dependent variable is what we are trying to forecast (i.e. sales, jobs, etc.). The independentvariable is what we are using to predict the forecast. For trend projection the independentvariable is always time; period 1, 2, 3, etc.

    For causal regression the sky is our limit with respect to independent variables. For example,we can predict season ticket sales for the Astros based on their previous seasons record. Orwe can predict our grade on final exam based on our grades on tests 1 and 2. As you can see,in casual regression, we can use any other variable to help us forecast. A great example is theinterest rate and new home sales! Lets look further.

  • 8/3/2019 Lecture.chap15

    27/32

    Slide 40

    40 2003 Thomson/South-Western Slide

    Trend Projection

    Using the method of least squares, the formula for thetrend projection is: Tt = b0 + b1t.

    where: Tt = trend forecast for time period t

    b1 = slope of the trend line

    b0 = trend line projection for time 0

    b1 = ntYt - t Yt

    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=

    Y

    t

    I have provided you the equations that are used to determine the least sum of squares trendprojection line. As you can see there are a lot of opportunities for error, so I suggest we useExcel.

  • 8/3/2019 Lecture.chap15

    28/32

    Slide 41

    41 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service

    The number of plumbing repair jobs performed byAuger's Plumbing Service in each of the last ninemonths are listed below.

    Month Jobs Month Jobs Month Jobs

    March 353 June 374 September 399

    April 387 July 396 October 412

    May 342 August 409 November 408

    Forecast the number of repair jobs Auger's will

    perform in December using the least squares method.

    We will use Augers Plumbing service to demonstrate trend projections. We will see how to useExcel to compute Trend projection.

    Slide 42

    42 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service

    Trend Projection (continued)

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

    ntYt - t Yt (9)(17844) - (45)(3480)b1 = = = 7.4

    nt 2 - (t)2 (9)(285) - (45)2

    = 386.667 - 7.4(5) = 349.667

    T10 = 349.667 + (7.4)(10) = 423.667

    0 1b Y b t=

    Yt

  • 8/3/2019 Lecture.chap15

    29/32

    Slide 43

    43 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service

    Excel Spreadsheet Showing Input Data

    A B C

    1 Auger's Plumbing Service

    2

    3 Month Calls

    4 1 353

    5 2 387

    6 3 342

    7 4 374

    8 5 396

    9 6 409

    10 7 399

    11 8 412

    12 9 408

    13

    Slide 44

    Slide shows how to set up in Excel.

    44 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service

    Steps to Trend Projection Using Excel

    Step 1: Select an empty cell (B13) in the worksheet.

    Step 2: Select the Insert pull-down menu.

    Step 3: Choose the Function option.

    Step 4: When the Paste Function dialog box appears:

    Choose Statistical in Function Categorybox.

    Choose Forecast in the Function Name box.

    Select OK.

    more . . . . . . .

  • 8/3/2019 Lecture.chap15

    30/32

    Slide 45

    Continue

    45 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service

    Steps to Trend Projecting Using Excel (continued)

    Step 5: When the Forecast dialog box appears:

    Enter 10 in the x box (for month 10).

    Enter B4:B12 in the Known ys box.

    Enter A4:A12 in the Known xs box.

    Select OK.

    Slide 46

    46 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service

    Spreadsheet with Trend Projection for Month 10

    A B C

    1 Auger's Plumbing Service

    2

    3 Month Cal ls

    4 1 353

    5 2 387

    6 3 342

    7 4 374

    8 5 396

    9 6 409

    10 7 399

    11 8 412

    12 9 408

    13 10 423.667 Projected

  • 8/3/2019 Lecture.chap15

    31/32

    Slide 47

    47 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service (B)

    Forecast for December (Month 10) using a

    three-period (n = 3) weighted moving average with

    weights of .6, .3, and .1.

    Then, compare this Month 10 weighted moving

    average forecast with the Month 10 trend projection

    forecast.

    Slide 48

    48 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service (B)

    Three-Month Weighted Moving Average

    The forecast for December will be the weightedaverage of the preceding three months: September,October, and November.

    F10 = .1YSep. + .3YOct. + .6YNov.= .1(399) + .3(412) + .6(408)

    =

    Trend Projection

    F10 = 423.7 (from earlier slide)

    408.3

  • 8/3/2019 Lecture.chap15

    32/32

    Slide 49

    49 2003 Thomson/South-Western Slide

    Example: Augers Plumbing Service (B)

    Conclusion

    Due to the positive trend component in the time

    series, the trend projection produced a forecast that is

    more in tune with the trend that exists. The weighted

    moving average, even with heavy (.6) placed on the

    current period, produced a forecast that is lagging

    behind the changing data.

    Slide 50

    50 2003 Thomson/South-Western Slide

    End of Chapter 15