Chapter 6 Demand Forecasting
Lecture plan
Meaning of Demand Forecasting
Techniques of Demand Forecasting
Subjective Methods of Demand Forecasting
Survey methods
Expert opinion methods
Quantitative Methods of Demand Forecasting
Trend methods
Smoothing methods
Simulation
Statistical methods
Limitations of Demand Forecasting
Objectives
To introduce the relevance of demand forecasting in business. To understand the types of demand forecasting. To explore qualitative techniques of forecasting demand. To understand quantitative and econometric methods of demand forecasting. To point out the limitations of demand forecasting.
Meaning of Demand Forecasting
“An estimate of sales in dollars or physical units for a specified future period under a proposed marketing plan.” American Marketing Association
Demand forecasting is the scientific and analytical estimation of demand for a product (service) for a particular period of time. It is the process of determining how much of what products is needed when and where. An operations research technique of planning and decision making.
Categorization of Demand Forecasting By Level of Forecasting Firm (Micro) level: forecasting of demand for its product by an individual firm. decisions related to production and marketing. Industry level: for a product in an industry as a whole. insight in growth pattern of the industry in identifying the life cycle stage of the product relative contribution of the industry in national income. Economy (Macro) level: forecasting of aggregate demand (or output) in the economy as a whole. helps in various policy formulations at government level.
Categorization of Demand Forecasting By nature of goods Capital Goods: Derived demand demand for capital goods depends upon demand of consumer goods which they can produce. Consumer Goods: Direct demand durable consumer goods: new demand or replacement demand Non durable consumer goods: FMCG By Time Period Short Term (0 to 3 months): for inventory management and scheduling. Medium Term (3 months to 2 years): for production planning, purchasing, and distribution. Long Term (2 years and more): may extend up to 10 to 20 years.
for capacity planning, facility location, and strategic planning, long term capital requirement, and investment decisions.
Choice of a forecasting technique
depends on:
Imminent objectives of forecast, whether it is for a new product, or to gauge impact of a new advertisement, etc. Cost involved, cost of forecasting should not be more than its benefits, here opportunity cost of resources will also be important. Time perspective, whether the forecast is meant for the short run or the long run Complexity of the technique, vis-à-vis availability of expertise; this would determine whether the firm would look for experts “in house” or outsource it Nature and quality of available data, i.e. does the time series show a clear trend or is it highly unstable.
Techniques of Demand Forecasting
Subjective (Qualitative) methods: rely on human judgment and opinion.
Buyers’ Opinion Sales Force Composite Market Simulation Test Marketing Experts’ Opinion Group Discussion Delphi Method
Quantitative methods: use mathematical or simulation models based on historical demand or relationships between variables. Trend Projection Smoothing Techniques Barometric techniques Econometric techniques
Subjective Methods of Demand Forecasting Consumers’ Opinion Survey
Buyers are asked about future buying intentions of products, brand preferences and quantities of purchase, response to an increase in the price, or an implied comparison with competitor’s products. Census Method: Involves contacting each and every buyer Sample Method: Involves only representative sample of buyers Merits
Simple to administer and comprehend. Suitable when no past data available. Suitable for short term decisions regarding product and promotion.
Demerits
Expensive both in terms of resources and time. Buyers may give incorrect responses. Investigators’ bias regarding choice of sample and questions cannot be fully eliminated.
Subjective Methods of Demand Forecasting
Contd…
Sales Force Composite
Salespersons are in direct contact with the customers. Salespersons are asked about estimated sales targets in their respective sales territories in a given period of time. Merits Cost effective as no additional cost is incurred on collection of data. Estimated figures are more reliable, as they are based on the notions of salespersons in direct contact with their customers. Demerits Results may be conditioned by the bias of optimism (or pessimism) of salespersons. Salespersons may be unaware of the economic environment of the business and may make wrong estimates. This method is ideal for short term and not for long term forecasting
Subjective Methods of Demand Forecasting Contd…
Experts’ Opinion Method i) Group Discussion: (developed by Osborn in 1953) Decisions may be taken with the help of brainstorming sessions or by structured discussions. ii) Delphi Technique: developed by the Rand Corporation at the beginning of the Cold War, to forecast impact of technology on warfare. Way of getting repeated opinion of experts without their face to face interaction. Consolidated opinions of experts is sent for revised views till conclusions converge on a point.
Merits Decisions are enriched with the experience of competent experts. Firm need not spend time, resources in collection of data by survey. Very useful when product is absolutely new to all the markets.
Demerits Experts’ may involve some amount of bias. With external experts, risk of loss of confidential information to rival firms.
Subjective Methods of Demand Forecasting Contd…..
Market Simulation
Firms create “artificial market”, consumers are instructed to shop with some money. “Laboratory experiment” ascertains consumers’ reactions to changes in price, packaging, and even location of the product in the shop.
Grabor-Granger test:
Half of members are shown new product to see whether they would actually buy it at various prices on a random price list and then are shown the existing product. Other half is shown the existing product first and then the new product to ascertain if a product would be bought at different prices.
Merits
Market experiments provide information on consumer behaviour regarding a change in any of the determinants of demand. Experiments are very useful in case of an absolutely new product.
Demerits
People behave differently when they are being observed. In Grabor-Granger tests consumers may not quote the price they may pay.
Subjective Methods of Demand Forecasting Contd….
Test Marketing
Involves real markets in which consumers actually buy a product without the consciousness of being observed. product is actually sold in certain segments of the market, regarded as the “test market”. Choice and number of test market(s) and duration of test are very crucial to the success of the results. Merits
Most reliable among qualitative methods. Very suitable for new products. Considered less risky than launching the product across a wide region.
Demerits
Very costly as it requires actual production of the product, and in event of failure of the product the entire cost of test is sunk. Time consuming to observe the actual buying pattern of consumers.. Extrapolation of figures for calculating demand in widely varying markets across its geographical regions may not give accurate results.
Quantitative Methods of Demand Forecasting Trend Projection Statistical tool to predict future values of a variable on the basis of time series data. Time series data are composed of:
Secular trend (T): change occurring consistently over a long time and is relatively smooth in its path. Seasonal trend (S): seasonal variations of the data within a year Cyclical trend (C): cyclical movement in the demand for a product that may have a tendency to recur in a few years Random events (R): have no trend of occurrence hence they create random variation in the series. Additive Form: Y = T + S + C + R………..(1) Multiplicative Form: Y = T.S.C.R………….(2) Log Y= log T + log S + log C + log R………….(3)
Quantitative Methods: Methods of Trend Projection
Graphical method Past values of the variable on vertical axis and time on horizontal axis and line is plotted. Movement of the series is assessed and future values of the variable are forecasted simple but provides a general indication and fails to predict future value of demand 200 180 Demand for mobiles (in lakhs)
Contd…
160 140 120 100 80 60 40 20 0 2001
2002
2003 Ye a r
2004
2005
Quantitative Methods: Methods of Trend Projection
Contd…
Least squares method
based on the minimization of squared deviations between the best fitting line and the original observations given. Estimates coefficients of a linear function. Y=a+bX where a =intercept and b =slope The normal equations: ΣY=na + bΣX ΣXY= aΣX+ bΣX2 Once the coefficients of the trend equation are estimated, we can easily project the trend for future periods. Solving the normal equations: a= Y − b X b=
Σ(Y − Y )( X − X )
∑(X − X )
2
Quantitative Methods: Methods of Trend Projection
Contd…
ARIMA method: also known as Box Jenkins method
considered to be the most sophisticated technique of forecasting as it combines moving average and auto regressive techniques. Stage One: trend in the series is removed with help of ‘differencing’, i.e. the difference between values at adjacent period of time. Stage Two: Various possible combinations are created on basis of: i. order of involvement of auto regressive terms; ii. the order of moving average terms iii. the number of differences of the original series. Combinations are selected which provide an adequate fit to the series.
Stage Three: Parameter estimation is done using Least Squares. Stage Four: ‘Goodness of fit’ is tested and if it is not a good fit then the whole process is repeated from Stage Two. Stage Five: Once a ‘good fit’ is attained, its coefficients can be used to forecast future demand.
Quantitative Methods : Smoothing Techniques
Moving Average: forecasts on the basis of demand values during the recent past. n
Dn=
∑D i =1
i
where Di= demand in the ith period, n= number of periods in the
n
moving average
Weighted Moving Average: forecast the future value of sales on the basis of weights given to the most recent observations. The formula for computing weighted moving average is given as: n
Dn=
∑w D i =1
i
i
where Di= demand in the ith period, wi= weight for the ith
period, n= number of periods in the moving average.
Quantitative Methods : Smoothing Techniques
Contd…
Exponential Smoothing: assign greater weights to the most recent data, in order to have a more realistic estimate of the fluctuations. Weights usually lay between zero and one Ft+1=aDt+(1-a)Ft where Dt+1= forecast for the next period, Dt=actual demand in the present period, Ft=previously determined forecast for the present period, and a=weighting factor, termed as smoothing constant.
New forecast equals old forecast plus an adjustment for the error that had occurred in the last forecast
Ft+1=aDt+ a(1-a)Dt-1+ a(1-a)2Dt-2+ a(1-a)3Dt-3+...+a(1-a)t-1D1+ a(1-a)2Dt-2+ a(1-a)tF1)
Ft+1 is thus a weighted average of all past observations.
The older the data, the smaller the weight.
Quantitative Methods : Barometric Techniques
Contd….
Barometric Technique alerts businesses to changes in the overall economic conditions. Helps in predicting future trends on the basis of index of relevant economic indicators especially when the past data do not show a clear tendency of movement in a particular direction. Indicators may be Leading indicators: economic series that typically go up or down ahead of other series
Coincident
indicators: move up or down simultaneously with the
level of economic activities Lagging
lag.
series : which moves with economic series after a time
Quantitative Methods Contd…..
Simple (or Bivariate) Regression Analysis:
deals with a single independent variable that determines the value of a dependent variable. Demand Function: D = a+bP, where b is negative. If we assume there is a linear relation between D and P, there may also be some random variation in this relation. Sum of Squared Errors (SSE) : a measure of the predictive accuracy Smaller the value of SSE, the more accurate is the regression equation.
Nonlinear Regression Analysis Log linear function log D =A + B log P + e where A and B are the parameters to be estimated and e represents errors or disturbances. Linear form of log linear function D* = a + b P* + e where D*= log D and P*=log P
Quantitative Methods
Contd…..
Multiple Regression Analysis: D = a1+a2.P+a3.A+e (where A = advertising expenditure incurred).
D^ = a^1 + a^2P + a^3A, (where a1, a2 and a3 are the parameters and e is the random error term (or disturbance), having zero mean).
Similar to simple regression analysis, multiple regression analysis would aim at estimation of the parameters a1, a2 and a3. Choose such values of the coefficients that would minimize the sum of squares of the deviations.
Quantitative Methods
Contd…
Problems Associated with Regression Analysis
Multicollinearity: when two or more explanatory variables in the regression model are found to be highly correlated the estimated coefficients may not be accurately determined.
Heteroscedasticity: Classical regression models assume that the variance of error terms is constant for all values of the independent variables in the model; i.e. variables are homoscedastic.
Specification errors: Omission of one or more of the independent variables, or when the functional form itself is wrongly constructed or estimate a demand function in linear form, though the function should have been nonlinear.
Identification problem: where the variables, like a demand supply model.
equations
have
common
Quantitative Methods Simultaneous Equations Method
Based on the fact that in any economic decision every variable influences every other variable. Incorporates mutual dependence among variables. It is a simultaneous and two way relationships, A typical simultaneous equation model may comprise of:
Endogenous variables: included in the model as dependent variables Exogenous variables: given from outside the model Structural equations: which seek to explain the relation between a particular endogenous variable and other variables Definitional equations: which specify relationships that are considered to be true by definition
Limitations of Demand Forecasting
Change in Fashion: Is an inevitable consequence of advancement of civilization. Results of demand forecasting have short lasting impacts especially in a dynamic business environment.
Consumers’ Psychology: Results of forecasting depend largely on consumers’ psychology, understanding which itself is difficult.
Uneconomical: Requires collection of data in huge volumes and their analysis, which may be too expensive for small firms to afford. Estimation process may take a lot of time, which may not be affordable.
Lack of Experienced Experts: Accurate forecasting necessitates experienced experts, who may not be easily available. Forecasting by less experienced individuals may lead to erroneous estimates.
Lack of Past Data: Requires past sales data, which may not be correctly available. Typical problem in case for a new product.
Summary
Forecasting is an operations research technique of planning and decision making; demand forecasting is the scientific and analytical estimation of demand for a product (service) for a particular period of time. Demand forecasting can be categorized on basis of: i. the level of forecasting, i.e. firm, industry and economy; ii. time period, i.e. short run and long run iii. nature of goods, i.e. capital and consumer goods. Techniques of demand forecasting depend upon information on three questions: a. What do people say? b. What do people do? c. What have people done? In consumers’ opinion survey buyers are asked about their future buying intentions of products, their brand preferences and quantities of purchase. Future demand level may also be ascertained by experts with the help of brainstorming or by structured discussions or even by discussing without face to face interaction. Demand forecasting may also be done by market experiments conducted under controlled or simulated conditions or in real markets in which consumers actually buy a product without the awareness of being observed.
Summary
Trend projection is a powerful statistical tool frequently used to predict future values of a variable on the basis of time series data. Most time series data have components like seasonal trend, cyclical trend, secular trend and random events. Trend projection can be done by graphical method, least square method and ARIMA (Box Jenkins) method Smoothing techniques are used when the time series data exhibit little trend or seasonal variations, but a great deal of irregular or random variation. The most popular smoothing methods include moving average, weighted moving average and exponential smoothing. In barometric forecasting we construct an index of relevant economic indicators and forecast future trends on the basis of these indicators. Econometric methods apply statistical tools on economic theories to estimate economic variables. Regression analysis relates a dependant variable to one or more independent variables in the form of a linear equation. Regression can be linear, nonlinear and multiple. Simultaneous equations method incorporates mutual dependence among variables.
