Forecasting Meaning Elements Steps Types of forecasting MKA13

Forecasting Meaning Elements Steps Types of forecasting MKA/13 1

Forecasting FORECAST: A statement about the future Used to help managers Plan the system Plan the use of the system MKA/13 2

Common Features Assumes causal system past ==> future Forecasts rarely perfect because of randomness I see that you will Forecasts more accurate for get an A this quarter groups vs. individuals Forecast accuracy decreases as time horizon increases MKA/13 3

Elements of a Good Forecast Timely Reliable ul f g in n a Accurate Written e M MKA/13 y s Ea to e s u 4

Steps in the Forecasting Process “The forecast” Step 6 Monitor the forecast Step 5 Prepare the forecast Step 4 Gather and analyze data Step 3 Select a forecasting technique Step 2 Establish a time horizon Step 1 Determine purpose of forecast MKA/13 5

Types of forecast i. Qualitative ii. Time series analysis iii. Causal relationship iv. Simulation MKA/13 6

Qualitative subjective; Judgmental, based on estimates and opinions Can be of many types such as: i. iii. iv. v. Grass roots Market research Panel consensus Historical analogy Delphi method MKA/13 7

Grass roots Forecast by adding successively from the bottom Derives a forecast by compiling input from those at the end of hierarchy who deal with what is being forecast. As for example: An overall sales forecast may be derived by combining inputs from each sales person who is closest to his territory. MKA/13 8

Market research Firms often hire outside companies that specialize in market research to conduct this type of forecast. Typically used to forecast long range and new product sales. MKA/13 9

Panel consensus Free open exchange at meeting. Idea is that two heads are better than one Group discussion will produce better forecast than any one individual Participants may be executive, salespeople and customers MKA/13 10

Historical analogy Where a forecast may be derived by using the history of a similar product Where an existing product or generic product could be used as a model. Example can be complementary or substitute product. Demand for CD is caused by DVD players. MKA/13 11

Delphi method Group of experts responds to questionnaires A moderator compiles results and formulates a new questionnaire and submitted again to the respondents As a results initiate learning process and no influence of group pressure. MKA/13 12

Time series analysis Tries to predict the future based on past data Such as collected six weeks sales data can be used to predict 7 th week sales Can be of i. iii. iv. v. Simple moving average Weighted moving average Simple exponential smoothing Exponential smoothing with trend Linear regression MKA/13 13

Guide to select appropriate method FM Amt of historical data Data pattern Forecast horizon Simple moving average 6 to 12 months stationary Shortmedium Weighted moving average 5 -10 observations do short Simple exponential smoothing do Stationary and short trend Exponential smoothing with trend do do Linear regression 10 -20 observations at Stationary, least 5 seasonality, observations/season trend do Short to medium 14 MKA/13

Which model you choose? Depends on Ø Time horizon to forecast Ø Data availability Ø Accuracy required Ø Size of forecasting budget Ø Availability of qualified personnel MKA/13 15

Simple moving average A time period containing a number of data points is averaged by dividing the sum of the points values by the number of points When demand fro product is neither growing nor declining and if it does not have seasonal characteristics, this model can be used. Ft =At-1 +At-2+At-3……+At-n /n Ft = forecast for the coming period At-1 = Actual occurrence for the past period At-2 =Actual occurrence two periods ago n= no of periods to be averaged MKA/13 16

Weighted moving average Moving average allows any weight to be placed on each element The sum of all weights equal 1 Ft =w 1 At-1 + w 2 At-2+ w 3 At-3……+ w n At-n M 1 m 2 m 3 m 4 100 90 105 95 ? F 5=. 40*95+. 3*105+. 20*90+. 1*100 =97. 5 MKA/13 17

Exponential smoothing Only three pieces of data are used such as The most recent forecast The actual demand that occurred for that forecast period Smoothing constant α F t =Ft-1 + α (At-1 –Ft-1) F t = the exponential smooth forecast for period t Ft-1=Exponentially smoothed forecast made for the prior period. At-1 = The actual demand in the prior period α = the desired response rate or smoothing constant MKA/13 19

Why exponential smoothing Because of four reasons v. Are surprisingly accurate v. Formulating the model is relatively easy v. Little computation is required v. The user can understand how the model works. MKA/13 20

Linear regression analysis The past data and future projections are assumed to fall about a straight line Linear regression line is of the form Y is the dependent variable, a is the y intercept b is the slope t is the independent variable Y Yt = a + bt MKA/13 0 1 2 3 4 5 t 24

Calculating a and b n (ty) - t y B(Slope) = n t 2 - ( t) 2 A Intercept = y - b t n MKA/13 25

Linear Trend Equation Example MKA/13 26

Linear Trend Calculation 5 (2499) - 15(812) 12495 -12180 b = = = 6. 3 5(55) - 225 275 -225 812 - 6. 3(15) a= = 143. 5 5 y = 143. 5 + 6. 3 t MKA/13 27

Example Sunrise baking company markets cakes through a chain of food stores. It has been experiencing over and underproduction because of forecasting errors. The following data are its demand in dozens of cakes for the past four weeks. Cakes are made for the following day; for example Sunday's cake production is for Monday’s sale…. the bakery is closed Saturday, so Friday’s production must satisfy demand for both Saturday and Sunday MKA/13 28

Example Day 4 weeks 3 weeks ago 2 weeks ago Last week Monday 2200 2400 2300 2400 Tuesday 2200 2100 2200 Wednesday 2300 2400 2300 2500 Thursday 1800 1900 1800 2000 Friday 1900 1800 2100 2000 2800 2700 3000 2900 Saturday Sunday MKA/13 29

Example Make a forecast for this week on the following basis i. Daily using a simple four week moving average ii. Daily using a weighted average of. 4, . 3, . 2, . 1 for the past four weeks iii. Sun rise is also planning its purchases of ingredients for bread production. If bread demand had been forecast for last week at 22000 loaves and only 21000 loaves were actually demanded, what would sunrise’s forecast be for this week using exponential smoothing with a=. 10 iv. suppose with the forecast made in c this week’s demand actually turns out to be 22500. what would be the new forecast be for the next week MKA/13 30

Causal relationship forecasting One occurrence causes another The rain causes the sale of rain gear If housing starts are known then sale of carpet forecasting is possible MKA/13

Forecasting using a causal relationship Year 1998 1999 2000 2001 2002 2003 2004 MKA/13 Housing Permits x 18 15 12 10 28 35 30 Sales sqr mtr y 13000 12000 11000 14000 16000 19000 17000

Y=a+bx a is y intercept b is slope= y 2 -y 1/x 2 -x 1 = 17000 -1000/30 -10 y = 7000+350 x if house permit is 26 y= 7000+350*26 is the forecast of next year. Book 1. xlsx MKA/13

Simulation Dynamic model usually computer based Allow the forecaster to make assumptions about the internal variables and external environment in the model Depending on the variable in the model forecaster may ask such question as what would happen to my forecast if price increased by 10% MKA/13