DemandForecasting

Unit3

Unit3
Structure
3.1 3.2 3.3 3.4 3.5 3.6 Introduction Objectives MeaningAndUsesOfDemandForecasting LevelOfDemandForecasting CriteriaForGoodDemandForecasting MethodsOfDemandForecasting SurveyMethods 3.6.1 StatisticalMethods 3.6.2 DemandForecastingForNewProducts SelfAssessmentQuestions 3.7 Summary TerminalQuestions AnswertoSAQs&TQs

DemandForecasting

3.1 Introduction
An important aspect of demand analysis from the management point of view is concerned with forecasting demand for products, either existing or new. Demand forecasting refers to an estimate of most likely future demand for product under given conditions. Such forecasts are of immense use in making decisions with regard to production, sales, investment, expansion, employmentofmanpoweretc.,bothintheshortrunaswellasinthelongrun.Forecastsaremadeat micro level and macro level. There are different methods of forecasts like survey methods and statistical methods generally for the existing products and for new products depending upon the nature,numberofmethodslikeevolutionaryapproachsubstituteapproach,growthcurveapproach etc. LearningObjectives: Afterstudyingthisunit,youshouldbeabletounderstandthefollowing

1. Makeacomprehensivestudyoftheconditionsunderwhichthefirmisoperating.
SikkimManipalUniversity

41

DemandForecasting

Unit3

2. Haveproperknowledgeofbothsurveymethodsandstatisticalmethods. 3. Adoptasuitablemethodtomakeaccurateforecasts. 4. Usecommonsensemainly,withoutrelyingtoomuchonanyoneofthemethodstoarriveatright

conclusions.

3.2 MeaningAndFeatures
Demandforecasting seeks to investigateand measure theforces that determine salesfor existing andnewproducts.Generally companies plantheir business productionor sales inanticipation of future demand. Hence forecasting future demand becomes important. In fact it is the very soul of good business because every business decision is based on some assumptions about the future whether right or wrong, implicit or explicit. The art of successful business lies in avoiding or minimizingtherisksinvolvedasfaraspossibleandfacetheuncertaintiesinamostbefittingmanner .ThusDemandForecastingreferstoanestimationofmostlikelyfuturedemandforaproduct undergivenconditions. Importantfeaturesofdemandforecasting Itisbasicallyaguessworkbutitisaneducatedandwellthoughtoutguesswork. Itisintermsofspecificquantities Itisundertakeninanuncertainatmosphere. Aforecastismadeforaspecificperiodoftimewhichwouldbesufficienttotakeadecisionand putitintoaction. Itisbasedonhistoricalinformationandthepastdata. Ittellsusonlytheapproximatedemandforaproductinthefuture. Itisbasedoncertainassumptions. Itcannotbe100%preciseasitdealswithfutureexpecteddemand

Demandforecastingisneededtoknowwhetherthedemandissubjecttocyclicalfluctuationsornot, sothattheproductionandinventorypolicies,etc,canbesuitablyformulated Demandforecastingisgenerallyassociatedwithforecastingsalesandmanipulatingdemand.Afirm can make use of thesalesforecasts made by the industry asapowerful toolforformulating sales policyandsalesstrategy.Theycanbecomeactionguidestoselectthecourseofactionwhichwill maximize thefirms earnings.When external economicfactors like the size of market, competitors
SikkimManipalUniversity

42

DemandForecasting

Unit3

attitudes,movementinprices,consumertastes,possibilitiesofnewthreatsfromsubstituteproducts etc, influence sales forecasting, internal factors like money spent on advertising, pricing policy, productimprovements,saleseffortsetc.,helpinmanipulatingdemand.Tousedemandforecasting inanactiveratherthanapassiveway,managementmustrecognizethedegreetowhichsalesarea resultnotonlyofexternaleconomicenvironmentbutalsooftheactionofthecompanyitself. Managerialusesofdemandforecasting: Intheshortrun: Demand forecasts for short periods are made on the assumption that the company has a given productioncapacityandtheperiodistooshorttochangetheexistingproductioncapacity.Generally itwouldbeoneyearperiod. Productionplanning:Ithelpsindeterminingthelevelofoutputatvariousperiodsandavoiding underoroverproduction. Helps toformulaterightpurchasepolicy: Ithelpsinbettermaterialmanagement,ofbuying inputsandcontrolitsinventorylevelwhichcutsdowncostofoperation. Helpstoframerealisticpricingpolicy:Arationalpricingpolicycanbeformulatedtosuitshort runandseasonalvariationsindemand. Sales forecasting: It helps the company to set realistic sales targets for each individual salesmanandforthecompanyasawhole. Helps in estimating short run financial requirements: It helps the company to plan the financesrequiredforachievingtheproductionandsalestargets.Thecompanywillbeableto raisetherequiredfinancewellinadvanceatreasonableratesofinterest. Reduce the dependence on chances: Thefirm would be able to plan its production properly andfacethechallengesofcompetitionefficiently. Helps to evolve a suitable labour policy: A proper sales and production policies help to determinetheexactnumberoflabourerstobeemployedintheshortrun. Inthelongrun: Longrunforecastingofprobabledemandforaproductofacompanyisgenerallyforaperiodof3 to5or10years.

1. Businessplanning:
SikkimManipalUniversity

43

DemandForecasting

Unit3

Ithelpstoplanexpansionoftheexistingunitoranewproductionunit.Capitalbudgetingofafirm isbasedonlongrundemandforecasting.

2. Financialplanning:
Ithelpstoplanlongrunfinancialrequirementsandinvestmentprogramsbyfloatingsharesand debenturesintheopenmarket.

3. Manpowerplanning:
Ithelpsinpreparinglongtermplanningforimpartingtrainingtotheexistingstaffandrecruit skilledandefficientlabourforceforitslongrungrowth.

4. Businesscontrol:
Effectivecontrolovertotalcostsandrevenuesofacompanyhelpstodeterminethevalueand volumeofbusiness.Thisinitsturnhelpstoestimatethetotalprofitsofthefirm.Thusitis possibletoregulatebusinesseffectivelytomeetthechallengesofthemarket.

5. Determinationofthegrowthrateofthefirm:
Asteadyandwellconceiveddemandforecastingdeterminethespeedatwhichthecompany cangrow.

6. Establishmentofstabilityintheworkingofthefirm:
Fluctuationsinproductioncauseupsanddownsinbusinesswhichretardssmooth functioningofthefirm.Demandforecastingreducesproductionuncertaintiesandhelpin stabilizingtheactivitiesofthefirm.

7. Indicatesinterdependenceofdifferentindustries:
Demandforecastsofparticularproductsbecomethebasisfordemandforecastsofother relatedindustries,e.g.,demandforecastforcottontextileindustrysupplyinformationtothe mostlikelydemandfortextilemachinery,colour,dyestuffindustryetc.,

8. Moreusefulincaseofdevelopednations:
It is of great use in industrially advanced countries where demand conditions fluctuate much morethansupplyconditions. The above analysis clearly indicates the significance of demand forecasting in the modern businesssetup.

3.3 LevelsOfDemandForecasting
Demandforecastingmaybeundertakenatthreedifferentlevels,viz.,microlevelorfirmlevel, industrylevelandmacrolevel.
SikkimManipalUniversity

44

DemandForecasting

Unit3

Microlevelorfirmlevel Thisreferstothedemandforecastingbythefirmforitsproduct.Themanagementofafirmisreally interestedinsuchforecasting.Generallyspeaking,demandforecastingreferstotheforecastingof demandofafirm. Industrylevel Demandforecastingfortheproductofanindustryasawholeisgenerallyundertakenbythetrade associationsandtheresultsaremadeavailabletothemembers.Amemberfirmbyusingsuchdata andinformationmaydetermineitsmarketshare. Macrolevel Estimatingindustrydemandfortheeconomyasawholewillbebasedonmacroeconomicvariables like national income, national expenditure, consumption function, index of industrial production, aggregate demand, aggregate supply etc, Generally, it is undertaken by national institutes, govt. agenciesetc.SuchforecastsarehelpfultotheGovernmentindeterminingthevolumeofexportsand imports,controlofpricesetc. The managerial economist has to take into consideration the estimates of aggregate demand and alsoindustrydemandwhilemakingthedemandforecastfortheproductofaparticularfirm.

3.4 CriteriaForGoodDemandForecasting
Apartfrombeingtechnicallyefficientandeconomicallyidealagoodmethodofdemandforecasting shouldsatisfyafewbroadeconomiccriteria.Theyareasfollows:

1. Accuracy: Accuracy is the most important criterion of a demand forecast, even though cent
percentaccuracyaboutthefuturedemandcannotbeassured.Itisgenerallymeasuredinterms ofthepastforecastsonthepresentsalesandbythenumberoftimesitiscorrect.

2. Plausibility: The techniques used and the assumptions made should be intelligible to the
management.Itisessentialforacorrectinterpretationoftheresults.

SikkimManipalUniversity

45

DemandForecasting

Unit3

3. Simplicity:Itshouldbesimple,reasonableandconsistentwiththeexistingknowledge.Asimple
methodisalwaysmorecomprehensivethanthecomplicatedone

4. Durability: Durability of demand forecast depends on the relationships of the variables

consideredandthestabilityunderlyingsuchrelationships,asforinstance,therelationbetween price and demand, between advertisement and sales, between the level of income and the volumeofsales,andsoon.

5. Flexibility:Thereshouldbescopeforadjustmentstomeetthechangingconditions.Thisimparts
durabilitytothetechnique.

6. Availabilityofdata:Immediateavailabilityofrequireddataisofvitalimportancetobusiness.It
shouldbemadeavailableonanuptodatebasis.Thereshouldbescopeformakingchangesin thedemandrelationshipsastheyoccur.

7. Economy:Itshouldinvolvelessercostsasfaraspossible.Itscostsmustbecomparedagainst
thebenefitsofforecasts

8. Quickness:Itshouldbecapableofyieldingquickandusefulresults.Thishelpsthemanagement
totakequickandeffectivedecisions. Thus, an ideal forecasting method should be accurate, plausible, durable, flexible, make the data availablereadily,economicalandquickinyieldingresults.

3.5MethodsOrTechniquesOfForecasting
Demandforecastingisahighlycomplicatedprocessasitdealswiththeestimationoffuturedemand. Itrequirestheassistanceandopinionofexpertsinthefieldofsalesmanagement.Whileestimating futuredemand,oneshouldnotgivetoomuchofimportancetoeitherstatisticalinformation,pastdata or experience, intelligence and judgment of the experts. Demand forecasting, to become more realistic should consider the two aspects in a balanced manner. Application of commonsense is neededtofollowapragmaticapproachindemandforecasting. Broadlyspeaking,therearetwomethodsofdemandforecasting.Theyare:1.Surveymethodsand2 Statisticalmethods.

SikkimManipalUniversity

46

DemandForecasting

Unit3

METHODSOFDEMANDFORECASTING

SURVEYMETHODS

STATISTICALMETHODS

1.ConsumersinterviewMethod a..Surveyofbuyersintentions throughquestionnaire b.DirectInterviewMethod i..CompleteEnumerationMethod ii. SampleSurveyMethod 2.CollectiveOpinionMethod 3.ExpertOpinionmethod 4.EndUseMethod

1.Trendprojectionmethod 2.EconomicIndicators

3.6SurveyMethods
Surveymethodshelpusinobtaininginformationaboutthefuturepurchaseplansofpotentialbuyers through collecting the opinions of experts or by interviewing the consumers. These methods are extensively used in short run and estimating the demand for new products. There are different approachesundersurveymethods.Theyare A.Consumersinterviewmethod: Underthismethod,effortsaremadetocollecttherelevant informationdirectlyfromtheconsumerswithregardtotheirfuturepurchaseplans.Inorderto gatherinformationfromconsumers,anumberofalternativetechniquesaredevelopedfromtimeto time.Amongthem,thefollowingaresomeoftheimportantones. Survey of buyers intentions or preferences: It is one of the oldest methods of demand forecasting.ItisalsocalledasOpinionsurveys.

SikkimManipalUniversity

47

DemandForecasting

Unit3

Under this method, consumerbuyers are requested to indicate their preferences and willingness about particular products. They are asked to reveal their future purchase plans with respect to specific items. They are expected to give answers to questions like what items theyintendtobuy,inwhatquantity,why,where,when,whatqualitytheyexpect,howmuchmoney theyareplanningtospendetc.Generally,thefieldsurveyisconductedbythemarketingresearch department of the company or hiring the services of outside research organizations consisting of learnedandhighlyqualifiedprofessionals. The heart of the survey is questionnaire. It is a comprehensive one covering almost all questions either directly or indirectly in a most intelligent manner. It is prepared by an expert body who are specialistsinthefieldormarketing. Thequestionnaireisdistributedamongtheconsumerbuyerseitherthroughmailorinpersonbythe company.Consumersarerequestedtofurnishallrelevantandcorrectinformation. Thenextstepistocollectthequestionnairefromtheconsumersforthepurposeofevaluation.The materialscollectedwillbeclassified,editedanalyzed.Ifanybiasprejudices,exaggerations,artificial orexcessdemandcreationetc.,arefoundatthetimeofansweringtheywouldbeeliminated. Theinformationsocollectedwillnowbeconsolidatedandreviewedbythetopexecutiveswithlotof experience. It will be examined thoroughly. Inferences are drawn and conclusions are arrived at. Finallyareportispreparedandsubmittedtomanagementfortakingfinaldecisions. Thesuccessofthesurveymethoddependsonmanyfactors.1)Thenatureofthequestionsasked, 2) The ability of the surveyed 3) The representative of the samples 4) Nature of the product 5) characteristics of the market 6) consumer buyers behavior, their intentions, attitudes, thoughts, motives,honestyetc.7)Techniquesofanalysis8)conclusionsdrawnetc. The management should not entirely depend on the results of survey reports to project future demand.Consumerbuyersmaynotexpresstheirhonestandrealviewsandassuchtheymaygive onlythebroadtrendsinthemarket.Inordertoarriveatrightconclusions,fieldsurveysshouldbe regularlycheckedandsupervised. Thismethodissimpleandusefultotheproducerswhoproducegoodsinbulk.Heretheburdenof forecastingisputoncustomers. Howeverthismethodisnotmuchusefulinestimatingthefuturedemandofthehouseholdsasthey runinlargenumbersandalsodonotfreelyexpresstheirfuturedemandrequirements.Itisexpensive

SikkimManipalUniversity

48

DemandForecasting

Unit3

andalsodifficult.Preparationofaquestionnaireisnotaneasytask.Atbestitcanbeusedforshort termforecasting. B.DirectInterviewMethod Experience has shown that many customers do not respond to questionnaire addressed to them even if it is simple due to varied reasons. Hence, analternative method is developed. Under this method, customers are directly contacted and interviewed. Direct and simple questions are asked to them. They are requested to answer specifically about their budget, expenditure plans, particularitemstobeselected,thequalityandquantityofproducts,relativepricepreferencesetc.for aparticularperiodoftime.Therearetwodifferentmethodsofdirectpersonalinterviews.Theyareas follows: i.Completeenumerationmethod Underthismethod,allpotentialcustomersareinterviewedinaparticularcityoraregion.The answers elicited are consolidated and carefully studied to obtain the most probable demand for a product.Themanagementcansafelyprojectthefuturedemandforitsproducts.Thismethodisfree from all types of prejudices. The result mainly depends on the nature of questions asked and answersreceivedfromthecustomers. However,thismethodcannotbeusedsuccessfullybyallsellersinallcases.Thismethodcan beemployedtoonlythoseproductswhosecustomersareconcentratedinasmallregionorlocality. Incaseconsumersarewidelydispersed,thismethodmaynotbephysicallyadoptedorprovecostly bothintermsoftimeandmoney.Hence,thismethodishighlycumbersomeinnature. ii.Samplesurveymethodortheconsumerpanelmethod Experience of the experts show that it is impossible to approach all customers as such careful samplingofrepresentativecustomersisessential.Hence,anothervariantofcompleteenumeration method has been developed, which is popularly known as sample survey method. Under this method, different cross sections of customers that make up the bulk of the market are carefully chosen. Only such consumers selected from the relevant market through some samplingmethodareinterviewedorsurveyed.Inotherwords,agroupofconsumersarechosen and queried about their preferences in concrete situations. The selection of a few customers is known as sampling. The selected consumers form a panel. This method uses either random
SikkimManipalUniversity

49

DemandForecasting

Unit3

sampling or the stratified sampling technique. The method of survey may be direct interview or mailed questionnaire to the selected consumers. On the basis of the views expressed by these selectedconsumers,mostlikelydemandmaybeestimated.Theadvantageofapanelliesinthefact that the same panel is continued and new expensive panel does not have to be formulated every timeanewproductisinvestigated. Ascomparedtothecompleteenumerationmethod,thesamplesurveymethodislesstedious,less expensive,muchsimplerandlesstimeconsuming.Thismethodisgenerallyusedtoestimateshort rundemandbygovernmentdepartmentsandbusinessfirms. Successofthismethoddependsuponthesincerecooperationoftheselectedcustomers.Hence, selectionofsuitableconsumersforthespecificpurposeisofgreatimportance. Evenwithcarefulselectionofcustomersandthetruthfulinformationabouttheirbuyingintention,the resultsofthesurveycanonlybeoflimiteduse.Asuddenchangeinprice,inconsistencyinbuying intentions of consumers, number of sensible questions asked and dropouts from the panel for variousreasonsputaseriouslimitationonthepracticalusefulnessofthepanelmethod. C.Collectiveopinionmethodoropinionsurveymethod This is a variant of the survey method. This method is also known as Sales force polling or Opinionpollmethod.Underthismethod,salesrepresentatives,professionalexpertsandthe market consultants and others are asked to express their considered opinions about the volumeofsalesexpectedinthefuture.Thelogicandreasoningbehindthemethodisthatthese salesmen and other people connected with the sales department are directly involved in the marketing and selling of the products in different regions. Salesmen, being very close to the customers,willbeinapositiontoknowandfeelthecustomersreactionstowardstheproduct.They canstudythepulseofthepeopleandidentifythespecificviewsofthecustomers.Thesepeopleare quite capable of estimating the likely demand for the products with the help of their intimate and friendly contact with the customers and their personal judgments based on the past experience. Thus, they provide approximate, if not accurate estimates. Then, the views of all salesmen are aggregatedtogettheoverallprobabledemandforaproduct. Further,theseopinionsorestimatescollectedfromthevariousexpertsareconsidered,consolidated and reviewed by the top executives to eliminate the bias or optimism and pessimism of different salesmen.Theserevisedestimatesarefurtherexaminedinthelightoffactorslikeproposedchange

SikkimManipalUniversity

50

DemandForecasting

Unit3

in selling prices,product designsand advertisement programs, expected changes in thedegree of competition,incomedistribution,populationetc.Thefinalsalesforecastwouldemergeafterthese factors have been taken into account. This method heavily depends on the collective wisdom of salesmen,departmentalheadsandthetopexecutives. Itissimple,lessexpensiveandusefulforshortrunforecastingparticularlyincaseofnewproducts. The main drawback is that it is subjective and depends on the intelligence and awareness of the salesmen.Itcannotberelieduponforlongtermbusinessplanning. D.DelphiMethodorExpertsOpinionMethod ThismethodwasoriginallydevelopedatRandCorporationinthelate1940sbyOlafHelmer,Dalkey and Gordon. This method was used to predict future technological changes. It has proved more usefulandpopularinforecastingnoneconomicratherthaneconomicalvariables. It is a variant of opinion poll and survey method of demand forecasting. Under this method, outsideexpertsareappointed.Theyaresuppliedwithallkindsofinformationandstatistical data.Themanagementrequeststheexpertstoexpresstheirconsideredopinionsandviews about the expected future sales of the company. Their views are generally regarded as most objectiveones.TheirviewsgenerallyavoidorreducetheHaloEffectsandEgoInvolvement of the views of the others. Since experts opinions are more valuable, a firm will give lot of importancetothemandpreparetheirfutureplanonthebasisoftheforecastsmadebytheexperts. E.EndUseorInputOutputMethod Underthis method,thesaleoftheproductunderconsiderationisprojectedonthebasis of demand surveys of the industries using the given product as an intermediate product. The demand for the final product is the end use demand of the intermediate product used in the productionofthefinalproduct.Anintermediateproductmayhavemanyendusers,Fore.g.,steel can be usedfor making various types of agricultural and industrial machinery,for construction,for transportation etc. It may have the demand both in the domestic market as well as international market. Thus, end use demand estimation of an intermediate product may involve many final goods industries using this product, at home and abroad. Once we know the demand for final consumptiongoodsincludingtheirexportswecanestimatethedemandfortheproductwhichisused as intermediate good in the production of these final goods with the help of input output

SikkimManipalUniversity

51

DemandForecasting

Unit3

coefficients.Theinputoutputtablecontaininginputoutputcoefficientsforparticularperiodsare madeavailableineverycountryeitherbytheGovernmentorbyresearchorganizations. This method is used to forecast the demand for intermediate products only. It is quite useful for industries whichare largely producers goods, like aluminum, steel etc. The main limitation of the methodisthatasthenumberofendusersofaproductincrease,itbecomesmoreinconvenientto usethismethod.

3.6.1StatisticalMethod
Itisthesecondmostpopularmethodofdemandforecasting.Itisthebestavailabletechniqueand most commonly used method in recent years. Under this method, statistical, mathematical models,equationsetcareextensivelyusedinordertoestimatefuturedemandofaparticular product. They are used for estimating long term demand. They are highly complex and

complicated in nature. Some of them require considerable mathematical back ground and competence. Theyusehistoricaldatainestimatingfuturedemand.Theanalysisofthepastdemandservesasthe basisforpresenttrendsandbothofthembecomethebasisforcalculatingthefuturedemandofa commodityinquestionaftertakingintoaccountoflikelychangesinthefuture. There are several statistical methods and their application should be done by some one who is reasonablywellversedinthemethodsofstatisticalanalysisandintheinterpretationoftheresultsof suchanalysis. A.TrendProjectionMethod An old firm operating in the market for a long period will have the accumulated previous data on eitherproductionorsalespertainingtodifferentyears.Ifwearrangetheminchronologicalorder,we get what is called as time series. It is an ordered sequence of events over a period of time pertainingtocertainvariables.Itshowsaseriesofvaluesofadependentvariablesay,salesasit changesfromonepointoftimetoanother.Inshort,atimeseriesisasetofobservationstakenat specifiedtime,generallyatequalintervals.Itdepictsthehistoricalpatternundernormalconditions. This method is not based on any particular theory as to what causes the variables to change but merelyassumesthatwhateverforcescontributedtochangeintherecentpastwillcontinuetohave the same effect. On the basis of time series, it is possible to project the future sales of a company.
SikkimManipalUniversity

52

DemandForecasting

Unit3

Further, the statistics and information with regard to the sales call for further analysis. When we representthetimeseriesintheformofagraph,wegetacurve,thesalescurve.Itshowsthetrendin salesatdifferentperiodsoftime.Also,itindicatesfluctuationsandturningpointsindemand.Ifthe turningpointsarefewandtheirintervalsarealsowidelyspread,theyyieldacceptableresults.Here the time series show a persistent tendency to move in the same direction. Frequency in turning pointsindicatesuncertaindemandconditionsandinthiscase,thetrendprojectionbreaksdown. Themajortaskofafirmwhileestimatingthefuturedemandliesinthepredictionofturningpointsin thebusinessratherthanintheprojectionoftrends.Whenturningpointsoccurmorefrequently,the firm has to make radical changes in its basic policy with respect to future demand. It is for this reasonthattheexpertsgiveimportancetoidentificationofturningpointswhileprojectingthefuture demandforaproduct. Theheartofthismethodliesintheuseoftimeseries.Changesintimeseriesariseonaccountof thefollowingreasons:

1. 2.

Secularorlongrunmovements: Secularmovementsindicatethegeneralconditionsand directioninwhichgraphofatimeseriesmoveinrelativelyalongperiodoftime. Seasonal movements: Time series also undergo changes during seasonal sales of a company. During festival season, sales clearance season etc., we come across most unexpectedchanges.

3. 4.

Cyclical Movements: It implies change in time series or fluctuations in the demand for a productduringdifferentphasesofabusinesscyclelikedepression,revival,boometc. Randommovement.Whenchangestakeplaceatrandom,wecallthemirregularorrandom movements. These movements imply sporadic changes in time series occurring due to unforeseeneventssuchasfloods,strikes,elections,earthquakes,droughtsandothersuch naturalcalamities.Suchchangestakeplaceonlyintheshortrun.Stilltheyhavetheirown impactonthesalesofacompany.

Animportantquestioninthisconnectionishowtoascertainthetrendintimeseries?Astatistician,in ordertofindoutthepatternofchangeintimeseriesmaymakeuseofthefollowingmethods.

1. TheLeastSquaresmethod. 2. TheFreehandmethod. 3. Themovingaveragemethod.

SikkimManipalUniversity

53

DemandForecasting

Unit3

4. Themethodofsemiaverages.
The method of Least Squares is more scientific, popular and thus more commonly used when compared to the other methods. It uses the straight line equation Y= a + bx tofit the trend to the data. Illustration. Under this method, the past data of the company are taken into account to assess the nature of present demand. On the basis of this information, future demand is projected. For e.g., A businessmanwillcollectthedatapertainingtohissalesoverthelast5years.Thestatisticsregarding thepastsalesofthecompanyisgivenbelow. Thetableindicatesthatthesalesfluctuateoveraperiodof5years.However,thereisanuptrendin thebusiness.Thesamecanberepresentedinadiagram.

Diagrammaticrepresentation. a) DerivingsalesCurve.

Y
Year Sales (Rs.) 1990 1991 1992 1993 1994 30 40 35 50 45

50 45 Sales 40 35 30 0 9091 92 93 94 Years

Salescurve

Wecanfindoutthetrendvaluesforeachofthe5yearsandalsoforthesubsequentyears makinguseofastatisticalequation,themethodofLeastSquares.Inatimeseries,xdenotestime andydenotesvariable.Withthepassageoftime,weneedtofindoutthevalueofthevariable. Tocalculatethetrendvaluesi.e.,Yc,theregressionequationusedis

SikkimManipalUniversity

54

DemandForecasting

Unit3

Yc =a+bx. As the values of a and b are unknown, we can solve thefollowing twonormal equations simultaneously.

(i)

Y=Na+bx
2 XY=ax+bx

(ii)

Where, Y=Totaloftheoriginalvalueofsales(y) N=Numberofyears, X=totalofthedeviationsoftheyearstakenfromacentralperiod. XY=totaloftheproductsofthedeviationsofyearsandcorrespondingsales(y)

2 X =totalofthesquareddeviationsofXvalues.

WhenthetotalvaluesofX.i.e., X=0

Year=n Salesin Y 1990 1991 1992 1993 1994 N=5 30 40 35 50 45 Y=200

Deviation yearX 2 1 0 +1 +2 X=0

SquareofDeviation Productsales andtime DeviationXY +4 +1 0 +1 +4 X2 =10 60 40 0 +50 +90 XY=40

Computed trend valuesYc 32 36 40 44 48

2 RsLakhs fromassumed X

Regressionequation=Yc =a+bx Tofindthevalueofa= Y/N=200/5=40 Tofindoutthevalueofb= XY/ X2 =40/10=4

SikkimManipalUniversity

55

DemandForecasting

Unit3

For1990

Y=40+(4x2) Y=408=32

For1991

Y=40+(4x1) Y=404=36

For1992

Y=40+(40x0) Y=40+0=40

For1993

Y=40+(4X1) Y=40+4=44

For1994

Y=40+(4X2) Y=40+8=48

Forthenexttwoyears,theestimatedsaleswouldbe: For1995 Y=40+(4X3) Y=40+12=52 For1996 Y=40+(4X4) Y=40+16=56 FindingtrendvalueswhenEvenYearsaregiven.

Year=N SalesinRs lakhs=Y

Deviation FromAssumed year=X

Squareof Productsalesand Deviation =X

2

Computed trendvalues Yc

timedeviation=XY

1990 1991 1992

55 25 65

3 1 +1

9 1 1

165 25 +65

44 48 52

SikkimManipalUniversity

56

DemandForecasting

Unit3

1993 N=4

55 Y=200

+3 X=0

9
2 X =20

+165 XY=40

56

Note:

1. Whenevenyearsaregiven,thebaseyearwouldbeinbetweenthetwomiddleyears.Inthis
example,inbetweenthetwomiddleyearsis1991.5(oneyear=1whereas6months=.5)

2. Forthepurposeofsimplecalculation,weassumethevalueforeach6monthsi.e.o.5=1
Tofindoutthevalueofa=200/4=50 Tofindoutthevalueofb=40/20=2 a=50,b=2. Calculationforeachyear.Findingtrendvalues.

1991.5=BaseYearFor1990 90=3 90.5=2 91=1 91.5=0 92=+1For1992 92.5=+2 93=+3 For1993 For1991

Y=50+2X3 Y=506=44 Y=50+2X1 Y=502 =48 Y=50+2X1 Y=50+2=52 Y=50+2X3 Y=50+6=56

Derivingtrendline

SikkimManipalUniversity

57

DemandForecasting

Unit3

TrendLine

Sales

Salescurve

90 91

92 93 94 Years

Trendprojectionmethodrequiressimpleworkingknowledgeofstatistics,quiteinexpensiveand yieldsfairlyreliableestimatesoffuturecourseofdemand... While estimating future demand we assume that the past rate of change in the dependent variablewillcontinuetoremainthesameinfuturealso.Hence,themethodyieldsresultonlyforthat periodwhereweassumetherearenochanges.Itdoesnotexplainthevitalupturnsanddownturns insales,thusnotveryusefulinformulatingbusinesspolicies. B.EconomicIndicators Economicindicatorsasamethodofdemandforecastingaredevelopedrecently.Underthismethod, afeweconomicindicatorsbecomethebasisforforecastingthesalesofacompany.Aneconomic indicatorindicateschangeinthemagnitudeofaneconomicvariable.Itgivesthesignalabout thedirectionofchangeinaneconomicvariable. Thishelpsindecisionmakingprocessofa company.Wecanmentionafeweconomicindicatorsinthiscontext. 1.Constructioncontractssanctionedfordemandtowardsbuildingmaterialslikecement. 2. Personalincometowardsdemandforconsumergoods. 3. Agricultureincometowardsthedemandforagriculturalinputs,instruments,fertilizers,manure, etc, 4. 5. Automobileregistrationtowardsdemandforcarspareparts,petroletc., PersonalIncome,ConsumerPriceIndex,Moneysupplyetc.,towardsdemand Forconsumption goods.
SikkimManipalUniversity

58

DemandForecasting

Unit3

The above mentioned and other types of economic indicators are published by specialist organizationsliketheCentralStatisticalOrganizationetc.Theanalystshouldestablishrelationship between the sale of the product and the economic indicators to project the correct sales and to measureastowhatextenttheseindicatorsaffectthesales.Thejobofestablishingrelationshipisa highlydifficulttask.Thisisparticularlysoincaseofnewproductswheretherearenopastrecords. Underthismethod,demandforecastinginvolvesthefollowingsteps:

a.Theforecasterhastoensurewhetherarelationshipexistsbetweenthedemandfora productandcertainspecifiedeconomicindicators. b.Theforecasterhastoestablishtherelationshipthroughthemethodofleastsquareandderivethe regressionequation.Assumingtherelationshiptobelinear,theequation willbey=a+bx. c.Oncetheregressionequationisobtainedbyforecastingthevalueofx,economic indicatorcanbeappliedtoforecastthevaluesofY.i.e.demand. d. Pastrelationshipbetweendifferentfactorsmaynotberepeated.Therefore,thevalue judgmentisrequiredtoforecastthevalueoffuturedemand.Inadditiontoit,manyothernew factorsmayalsohavetobetakenintoconsideration. When economic indicators are used to forecast the demand, a firm should know whether the forecastingisundertakenforashortperiodorlongperiod.Itshouldcollectadequateandappropriate data and select the ideal method of demand forecasting. The next stage is to determine the most likelyrelationshipbetweenthedependentvariablesandfinallyinterprettheresultsoftheforecasting. However it is difficult to find out an appropriate economic indicator. This method is not useful in forecastingdemandfornewproducts.

3.6.2 DemandForecastingForANewProduct
Demandforecastingfornewproductsisquitedifferentfromthatforestablishedproducts.Herethe firms will not have any past experience or past data for this purpose. An intensive study of the economic and competitive characteristics of the product should be made to make efficient forecasts.

SikkimManipalUniversity

59

DemandForecasting

Unit3

ProfessorJoelDean,however,hassuggestedafewguidelinestomakeforecastingofdemandfor newproducts. a.Evolutionaryapproach Thedemandforthenewproductmaybeconsideredasanoutgrowthofanexistingproduct.Fore.g., Demand for new Tata Indica, which is a modified version of Old Indica can most effectively be projectedbasedonthesalesoftheoldIndica,thedemandfornewPulsorcanbeforecastedbased onthesalesoftheoldPulsor.Thuswhenanewproductisevolvedfromtheoldproduct,thedemand conditionsoftheoldproductcanbetakenasabasisforforecastingthedemandforthenewproduct. b.Substituteapproach Ifthenewproductdevelopedservesassubstitutefortheexistingproduct,thedemandforthenew product may be worked out on the basis of a market share. The growths of demand for all the products have to be worked out on thebasis ofintelligentforecastsfor independent variables that influencethedemandforthesubstitutes.Afterthat,aportionofthemarketcanbeslicedoutforthe newproduct.Fore.g.,Amopedasasubstituteforascooter,acellphoneasasubstituteforaland line.Insomecasespriceplaysanimportantroleinshapingfuturedemandfortheproduct. c.OpinionPollapproach Underthisapproachthepotentialbuyersaredirectlycontacted,orthroughtheuseofsamplesofthe newproductandtheirresponsesarefoundout.Thesearefinallyblownuptoforecastthedemandfor thenewproduct. d.Salesexperienceapproach Offer the new product for sale in a sample market say supermarkets or big bazaars in big cities, whicharealsobigmarketingcenters.Theproductmaybeofferedforsalethroughonesupermarket andtheestimateofsalesobtainedmaybeblownuptoarriveatestimateddemandfortheproduct. e.GrowthCurveapproach According to this, the rate of growth and the ultimate level of demand for the new product are estimatedonthebasisofthepatternofgrowthofestablishedproducts.Fore.g.,AnAutomobileCo., whileintroducinganewversionofacarwillstudythelevelofdemandfortheexistingcar. f.Vicariousapproach Afirmwillsurveyconsumersreactionstoanewproductindirectlythroughgettingintouchwithsome specialized and informed dealers who have good knowledge about the market, about the different

SikkimManipalUniversity

60

DemandForecasting

Unit3

varietiesoftheproductalreadyavailableinthemarket,theconsumerspreferencesetc.Thishelpsin makingamoreefficientestimationoffuturedemand. These methods are not mutually exclusive. The management can use a combination of several of themsupplementandcrosscheckeachother.

3.7Summary
An important aspect of demand analysis from the management point of view is concerned with forecastingdemandeitherforexistingornewproducts.Demandforecastingreferstotheestimation offuturedemandundergivenconditions.Suchforecastshaveimmensemanagerialusesintheshort run like production planning, formulating right purchase policy, pricing policy, sales forecasting, estimatingshortrunfinancialrequirements,reducingthedependenceonchances,evolvingsuitable laborpolicy,controlonstocksetc.Inthelongruntheyhelpinefficientbusinessplanning,financial planning,regulatingbusinessefficiently,determinationofgrowthrateoffirm,stabilizingtheactivities of the firm and help in the growth of industries dependent on each other providing required informationparticularlyinthedevelopednations.Demandforecastsaredoneatmicrolevel,industry levelandmacrolevel.Agooddemandforecastingmethodmustbeaccurate,plausible,economical, durable,flexible,simplequickyieldingandpermitchangesinthedemandrelationshipsonanupto datebasis. Broadly speaking thereare two methods ofdemandforecasting 1.Survey Methods, 2Statistical Methods. Under the survey methods there are a number of variants like consumers interview method, collective opinion method, experts opinion method and enduse method. Under the consumers interview method demandforecasting is doneeither by conductinga surveyofbuyers intentionsthroughquestionnaireorbyinterviewingdirectlyalltheconsumersresidinginaregionor by forming a panel of consumers. Under the collective opinion method forecasts are made on the basis of the information gathered from the sales men and market experts regarding the future demandfortheproduct.UndertheExpertopinionmethodassistanceofoutsideexpertsaretakento forecastfuturedemand.Theendusemethodisadoptedtoforecastthedemandfortheintermediate productsmakinguseoftheinputoutputcoefficientsforparticularperiods. Statisticalmethodsliketrendprojectionandeconomicindicatorsaregenerallyusedtomakelongrun demandforecasts.Underthetrendprojectionmethod,basedonthepastdata,adoptingaregression analysis demand forecasts are made. Sometimes changes in the magnitude of the economic

SikkimManipalUniversity

61

DemandForecasting

Unit3

variablestooserveasabasisfordemandforecasting.Ariseinthepersonalincomeindicatesarise inthedemandforconsumptiongoods. Incaseofnewproductsasthefirmwillnothaveanypastexperienceorpastsalesdata,itwillhave to follow a few guidelines while making demand forecasts. Depending upon the nature of the development of the product different approaches like evolutionary approach, substitute, growth curve,opinionpoll,salesexperience,vicariousetc.,areadopted. Thusanumberofmethodsarebeingadoptedtoestimatethefuturedemandfortheproducts,which isofverygreatimportanceintheefficientmanagementofthebusiness. SelfAssessmentQuestions1 1. Demandforecastingreferstoanestimateof__________________fortheproductundergiven condition. 2. Theheartofthesurveymethodis___________________. 3. Collectiveopinionmethodisalsoknownas___________________. 4. SamplesurveymethodofDemandforecastingisalsocalled_________. 5. Aneconomicindicatorchangesinthemagnitudeofan____________. 6. Onthebasesof___________________itispossibleofprojectfuturesalesofacompany. TerminalQuestions

1. WhatisDemandForecasting?Explain inbriefvariousmethodofforecastingdemand. 2. WhatisDemand forecasting.Explaintrendprojectionmethoddemandforecastingwith

illustration.

3. Explainthevariousmethodofdemandforecastingforanewproduct.
AnswerforSelfAssessmentQuestions1 1. Mostlikelyfuturedemand 2. Questionnaire 3. Salesforcepolling 4. Consumerpanel 5. Economicvariable 6. Timeseries

SikkimManipalUniversity

62

DemandForecasting

Unit3

AnswertoTerminalQuestions 1. Refertounit3.2,3.6 2. Refertounit3.2,3.7 3. Refertounit23

SikkimManipalUniversity

63