UNIVERSITY OF DHAKA

AssignmentOn

Methods of Demand Forecasting

Milton Kumar GuriaID: 3-13-25-076

Submission Date: 20.02.2016

Department of ManagementUniversity of Dhaka

All Rights Reserved

Submitted to: Prof. Dr. Md. Mosharraf Hossain

Index

Delphi Method……………………..………….………………………………………………………02

Trend Projection Method ………………………………………………………………………06

Moving Average Method…………..……………………………….…………………………...07

Chain Ratio Method…………………..….………………………….…………………………..11

Consumption level method…………..….……………………………………….……………...13

End User Method………………………..………………………….………………….………..14

References………………………………………………………………………………………..16

1 | P a g e

After gathering information about various aspects of the market and demand from primary and

secondary sources, an attempt may be made to estimate future demand. A wide range of forecasting

methods are available to the market analyst. The important ones are discussed below:

Delphi MethodThe Delphi method is a structured communication technique or method, originally developed as a

systematic, interactive forecasting method which relies on a panel of experts. The experts answer

questionnaires in two or more rounds. After each round, a facilitator or change agent provides an

anonymous summary of the experts’ forecasts from the previous round as well as the reasons they

provided for their judgments. Thus, experts are encouraged to revise their earlier answers in light of

the replies of other members of their panel. It is believed that during this process the range of the

answers will decrease and the group will converge towards the "correct" answer. Finally, the process

is stopped after a predefined stop criterion (e.g. number of rounds, achievement of consensus, and

stability of results and the mean or median scores of the final rounds determine the results.

Delphi is based on the principle that forecasts (or decisions) from a structured group of individuals

are more accurate than those from unstructured groups. The technique can also be adapted for use in

face-to-face meetings, and is then called mini-Delphi. Delphi has been widely used for business

forecasting and has certain advantages over another structured forecasting approach, prediction

markets.

Key characteristicsThe following key characteristics of the Delphi method help the participants to focus on the issues at hand and separate Delphi from other methodologies:

Structuring of information flow

The initial contributions from the experts are collected in the form of answers to questionnaires and their comments to these answers. The panel director controls the interactions among the participants by processing the information and filtering out irrelevant content. This avoids the negative effects of face-to-face panel discussions and solves the usual problems of group dynamics.2 | P a g e

Fig1.1: The Delphi Method communication structure

Regular feedback

Participants comment on their own forecasts, the responses of others and on the progress of the panel as a whole. At any moment they can revise their earlier statements. While in regular group meetings participants tend to stick to previously stated opinions and often conform too much to the group leader; the Delphi method prevents it.

Role of the facilitator

The person coordinating the Delphi method is usually known as a facilitator or Leader, and facilitates the responses of their panel of experts, who are selected for a reason, usually that they hold knowledge on an opinion or view. The facilitator sends out questionnaires, surveys etc. and if the panel of experts accept, they follow instructions and present their views. Responses are collected and analyzed, and then common and conflicting viewpoints are identified. If consensus is not reached, the process continues through thesis and antithesis, to gradually work towards synthesis, and building consensus.

3 | P a g e

Applications of Delphi MethodUse in forecasting

First applications of the Delphi method were in the field of science and technology forecasting. The objective of the method was to combine expert opinions on likelihood and expected development time, of the particular technology, in a single indicator. One of the first such reports, prepared in 1964 by Gordon and Helmer, assessed the direction of long-term trends in science and technology development, covering such topics as scientific breakthroughs, population control, automation, space progress, war prevention and weapon systems. Other forecasts of technology were dealing with vehicle-highway systems, industrial robots, intelligent internet, broadband connections, and technology in education.

Later the Delphi method was applied in other places, especially those related to public policy issues, such as economic trends, health and education. It was also applied successfully and with high accuracy in business forecasting.

The Delphi method has also been used as a tool to implement multi-stakeholder approaches for participative policy-making in developing countries. The governments of Latin America and the Caribbean have successfully used the Delphi method as an open-ended public-private sector approach to identify the most urgent challenges for their region ICT Action Plans. As a result, governments have widely acknowledged the value of collective intelligence from civil society, academic and private sector participants of the Delphi, especially in a field of rapid change, such as technology policies.

Use in policy-making

From the 1970s, the use of the Delphi technique in public policy-making introduces a number of methodological innovations. In particular:

The need to examine several types of items (not only forecasting items but, typically, issue items, goal items, and option items) leads to introducing different evaluation scales which are not used in the standard Delphi. These often include desirability, feasibility (technical and political) and probability, which the analysts can use to outline different scenarios: the desired scenario (from desirability), the potential scenario (from feasibility) and the expected scenario (from probability);

4 | P a g e

The complexity of the issues posed in public policy-making leads to give more importance to the arguments supporting the evaluations of the panelists; so these are often invited to list arguments for and against each option item, and sometimes they are given the possibility to suggest new items to be submitted to the panel;

For the same reason, the scaling methods, which are used to measure panel evaluations, often include more sophisticated approaches such as multi-dimensional scaling. Further innovations come from the use of computer-based (and later web-based) Delphi conferences.

According to Bolognini web-based Delphis offer two further possibilities, relevant in the context of interactive policy-making and e-democracy. These are:

The involvement of a large number of participants,

The use of two or more panels representing different groups (such as policy-makers, experts, citizens), which the administrator can give tasks reflecting their diverse roles and expertise, and make them to interact within ad hoc communication structures. For example, the policy community members (policy-makers and experts) may interact as part of the main conference panel, while they receive inputs from a virtual community (citizens, associations etc.) involved in a side conference. These web-based variable communication structures, which he calls Hyperdelphi (HD), are designed to make Delphi conferences "more fluid and adapted to the hyper textual and interactive nature of digital communication".

Advantages of Delphi Method

Simple to conduct Can be used where quantitative data is not available The forecast is reliable as it is based on the opinion of people who know the product very

well Takes less time Avoids direct confrontation of experts with one another ( no peer pressure, or extrinsic

pressure) Structured/organized group communication process ( condense experts opinions into a few

precise and clearly defined statements)   Cost effective and flexible/adaptable, fast, versatile. Both long and short term forecasting possible Useful for technology, new product and industry sales

5 | P a g e

Disadvantages of Delphi Method

Results are based on mere hunch of one or more persons and not on scientific analysis. The experts may be biased This method is subjective and the forecast could be unfavorably influenced by persons with

vested interests Difficulty of getting expert panel. Breakdown of forecast is not possible. No guidelines for determining consensus, sample size and sampling techniques Requires time/participant commitment Time delays between rounds in data collection process (multiple data collection, analysis ,

processing)  Concerns about the reliability of the technique

Trend Projection

Trend Projection: When numerical data are available, a trend can be plotted on graph paper to show

changes through time. If desired, the trend line can then be extended or "projected" into the future on

the basis of the recent rate of change. Such a projection shows where the trend should be at some

point in the future assuming there is no shift in the rate of change. Example: A population with a

steady 2 percent rate of annual growth will double in about thirty-five years.

Trend forecasting is a complicated but useful way to look at past sales or market growth, determine

possible trends from that data and use the information to extrapolate (conclude) what could happen in

the future. Marketing experts typically utilize trend forecasting to help determine potential future

sales growth. Many areas of a business can use forecasting, and examining the concept as it relates to

sales can help you gain an understanding of this tool.

Applications of Trend analysis Trend analysis is used to forecast market trends, sales growth, inventory levels and interest

rates.6 | P a g e

Trend analysis offers a measurable and verifiable method for businesses to project future

outcomes.

It can be used for failure analysis and as an early warning indicator of impending problems.

Where accurate historical information exists and valid relationships between variables can be

established, trend analysis is a precise tool for anticipating events.

Advantages of Trend Projection

Very simple method

Provides reasonably accurate forecasts

Quick and inexpensive

A trend analysis can be replicated, checked, updated and refined when necessary.

Disadvantages of Trend Projection

Can be used only if past data is available

It is not necessary that past trend may continue to hold good in future as well

Historical data may not give a true picture of an underlying trend. An obvious event like

hurricanes Katrina and Sandy will distort a normal business trend line, while others are more

subtle.

A major problem in forecasting trends involves identifying turning points.

Long-term projections need more data to support them, and that may not always be available, particularly for a new business or product line.

Moving Average

A moving average (rolling average or running average) is a calculation to analyze data points by

creating series of averages of different subsets of the full data set. It is also called a moving mean

(MM)

Given a series of numbers and a fixed subset size, the first element of the moving average is obtained

by taking the average of the initial fixed subset of the number series. Then the subset is modified by

7 | P a g e

"shifting forward"; that is, excluding the first number of the series and including the next number

following the original subset in the series. This creates a new subset of numbers, which is averaged.

This process is repeated over the entire data series. The plot line connecting all the (fixed) averages is

the moving average. A moving average is a set of numbers, each of which is the  average of the

corresponding subset of a larger set of datum points. A moving average may also use unequal weights

for each datum value in the subset to emphasize particular values in the subset.

Types of Moving Averages:

There are a lot of different moving averages that can be used. Six (6) Main Types of Moving Averages are as below:

1. Simple Moving Average (SMA)

2. Exponential Moving Average (EMA)

3. Weighted Moving Average (WMA)

4. Smoothed Moving Average (SMMA)

5. Triangular Moving Average (TMA)

6. Moving Average Envelopes

As a Simple Moving Average (SMA) example, consider a security with the following closing

prices over 15 days:

Week 1 (5 days) – 20, 22, 24, 25, 23

Week 2 (5 days) – 26, 28, 26, 29, 27

Week 3 (5 days) – 28, 30, 27, 29, 28

A 10-day MA would average out the closing prices for the first 10 days as the first data point. The

next data point would drop the earliest price, add the price on day 11 and take the average, and so on

as shown below.

As noted earlier, MAs lag current price action because they are based on past prices; the longer the

time period for the MA, the greater the lag. Thus a 200-day MA will have a much greater degree of

lag than a 20-day MA because it contains prices for the past 200 days. The length of the MA to use

depends on the trading objectives, with shorter MAs used for short-term trading and longer-term MAs

more suited for long-term investors. The 200-day MA is widely followed by investors and traders,

with breaks above and below this moving average considered to be important trading signals.

8 | P a g e

Basic Features of Moving Averages:

(1) Easy-to-use technical analysis tool, easy to assess results

(2) Signal trades in any financial market or asset (check crossovers below)

(3) Can be used in any timeframe (intraday, daily, weekly, monthly)

(4) Can confirm (or not) the results of other technical analysis tools (ADX, RSI etc)

(5) A moving average can be the sub-components of another technical analysis tool (for example RSI or MACD)

(6) Can draw beautiful charts or channels (check Moving Average Envelopes below)

(7) Eliminates the impact of Market Noise9 | P a g e

(8) Can be used as components of automated trading systems (for example Forex Robots)

Uses of Moving Average Method

A moving average is commonly used with time series data to smooth out short-term fluctuations and highlight longer-term trends or cycles. The threshold between short-term and long-term depends on the application, and the parameters of the moving average will be set accordingly. For example, it is often used in technical analysis of financial data, like stock prices, returns or trading volumes. It is also used in economics to examine gross domestic product, employment or other macroeconomic time series.

Advantages of Moving Average Method

Easily understood

Easily computed

Provides stable forecasts

Disadvantages of Moving Average Method

Requires saving lots of past data points: at least the N periods used in the moving average computation

Lags behind a trend

Ignores complex relationships in data

Casual methods for forecastingMore analytical than the preceding methods, causal methods seek to develop forecasts on the basis of cause – effect relationships specified in an explicit, quantitative manner. The important methods under this category are as follows:

Chain ratio method: A simple analytical approach, this method calls for applying a series of factors for developing a demand forecast.

Consumption level method: Useful for a product that is directly consumed; this method estimates consumption level on the basis of elasticity coefficients, the important ones being the income elasticity of demand and the price elasticity of demand.

10 | P a g e

End use method: Suitable for intermediate products, the end use method develops demand forecasts on the basis of the consumptions coefficient of the product for various uses.

Chain Ratio MethodChain Ratio Method: A simple analytical approach, this method calls for applying a series of factors for developing a demand forecast.An estimation method used by multiplying a base number by a chain of related percentages. It is frequently used in marketing to estimate the size of a target market.

Here is a step-by-step example of how one might use the Chain Ratio Method to estimate the number of TVs in the United States (numbers for illustration only):

Question: How many TVs are in the US?

Method:

1. There are the roughly 300 million people in the US (This is the base number).

2. There are roughly 4 people per household: 300,000,000/4 = 75,000,000 households.

3. There are roughly 2 TVs per household: 75,000,000 * 2 = 150,000,000 TVs.

Answer:

150 million TVs.

Chain Ratio Method for Calculating Market Potential

The chain ratio method kind of assumes that what happens for one market will happen proportionally in another market. So if we have information on sales or market share in one market, we can extrapolate those results to another market. One common application of the chain ratio method is where a local or regional company wants to calculate potential sales nationally. In this case, the chain ratio method would follow several steps.

i. First, identify the demographic characteristics of the customer groups to whom the analysis applies. Because the chain ratio method relies heavily on secondary data, often population data from the Census Bureau, it’s best to describe markets in terms of sex, age, and income. Although we may use more complex demographic descriptions for other purposes, data for

11 | P a g e

the chain ratio method will be more available, timely, and accurate if the demographic description is kept simple. Sometimes data giving estimates of population size by occupation and some psychographic variables are available. If the data are available and reliable, then they can be used. If not, keep the market description simple.

For example suppose a brand of expensive upscale cookware produced in California sells primarily to women between the ages of thirty-five and sixty-four with household incomes above $75,000 annually. This demographic group is simply defined and does not include extra variables such as marital status, family size, education, or occupation. Suppose the company sells through retailers up and down the West Coast.

ii. Second, estimate how many current customers your brand has in the target demographic group. If the group is relatively new, the estimate may be more tenuous. If your company is seeking to sell to the same demographic group but in another area, then the information ought to be easy find internally.

For example, suppose the upscale cookware company had ten thousand purchases by people in that demographic group. How does the company get access to that kind of information? Well, it could survey its dealers or distributors if it uses traditional retail channels. If it uses primarily online distribution, it could survey purchasers as they buy. Or it could rely on syndicated survey or panel data provided by a commercial marketing research provider. No matter what the source, every company with even a modestly sophisticated understanding of its customers ought to have access to information that describes its customers.

iii. Third, divide total sales of your brand to members of the target demographic and then divide by the number of customers in the target demographic to obtain the sales per customer.

For example suppose that the company estimated that it sold five million dollars’ worth of cookware to the target demographic group. Assuming that the ten thousand purchases were made by separate individuals, the company would estimate the average purchase per customer to be $500 (5,000,000 ÷ 10,000).

iv. Fourth, estimate the number of people in the target group that live in the area to which your brand plans to expand and then multiply that number of people by the average sales per customer in that group calculated in the previous step.

For example, suppose the cookware company wants to expand into Arizona, Utah, and Colorado. Using Census Bureau data, the company learns that in these three states there are approximately 406,000 women between thirty-five and sixty-four living in households with annual incomes above $75,000. The market potential calculation is very straightforward. Multiply the population of the

12 | P a g e

target audience in the new location by the amount spent per year by current customers of the same demographic description. Thus, 406,000 × 500 =$203,000,000.

To emphasize again, this figure is not the sales forecast; it is not a prediction of what they will sell. The market potential estimate is the amount that could possibly be sold if all members of the target audience decided to buy.

Advantage of the chain ratio method

The main advantage of the chain ratio method is its simplicity.

The data are usually readily available and the mathematics is not complicated at all.

Disadvantage of the chain ratio method

The biggest drawback to using the chain ratio method is its main underlying assumptions, which is that the same average purchase rate will hold from one area to another. Of course, companies have little way of knowing that. However, having an empirically based estimate as a starting point for market expansion planning is better than simply making those decisions from intuition alone.

Consumption Level MethodConsumption level method: Useful for a product that is directly consumed; this method estimates consumption level on the basis of elasticity coefficients, the important ones being the income elasticity of demand and the price elasticity of demand.

Income Elasticity: This reflects the responsiveness of demand to variations in income. It is calculated as:

E1 = [Q2 - Q1/ I2- I1] * [I1+I2/ Q2 +Q1]

Where

E1 = Income elasticity of demandQ1 = quantity demanded in the base yearQ2 = quantity demanded in the following yearI1 = income level in the base year I2 = income level in the following year

Price Elasticity: This reflects the responsiveness of demand to variations in price.

13 | P a g e

It is calculated as:

EP = [Q2 - Q1/ P2- P1] * [P1+P2/ Q2 +Q1]

Where EP = Price elasticity of demand Q1 = quantity demanded in the base year

Q2 = quantity demanded in the following year

P1 = price level in the base year

P2 = price level in the following year

End-use MethodEnd use method: Suitable for intermediate products, the end use method develops demand forecasts on the basis of the consumptions coefficient of the product for various uses. Example: Milk is a commodity which can be used as an intermediary good for the production of ice cream and other dairy products. Cement may be used for constructing houses, hotels, bridges, roads, etc.

End use method involves following steps –

1. Identifying the possible uses of product2. Defying the consumption co-efficient of the product for various uses.3. Projecting the output level for consuming industries4. Deriving the demand for the product.

This method may be illustrated with an example, A certain industrial chemical, Indchem, is used by four industries, Alpha, Beta, Kappa, and Gamma. The consumption coefficients for those industries, the projected output levels for these industries for the year X, and the projected demand for Indchem are shown in exhibition 1.2

Fig 1.2: Projected demand for Indchem are shown in exhibition

14 | P a g e

As is clear from the foregoing discussion, the key inputs required for the application of end-use method are:

1. Projected output levels of consuming industries (units) and2. Consumption coefficients.

It may be difficult to estimate the projected output levels of consuming (units). More important, the consumption coefficients may vary from one period to another in wake of technological changes and improvements in the methods of manufacturing. Hence the end-use method should be used judiciously.

Advantages of Consumers End Use method:

Yields accurate predictions

Provides sector wise demand forecast for different industries

Especially useful for producers goods.

Disadvantages of Consumers End Use method:

Requires complex and diverse calculations

Costlier

Time consuming

Industry data may not be readily available

15 | P a g e

References

1.  Dalkey, Norman; Helmer, Olaf (1963). "An Experimental Application of the Delphi Method to the use of experts". Management Science 9 (3): 458–467. doi:10.1287/mnsc.9.3.458.

2. Bernice B. Brown (1968). "Delphi Process: A Methodology Used for the Elicitation of Opinions of Experts.": An earlier paper published by RAND (Document No: P-3925, 1968, 15 pages)

3. Sackman, H. (1974), "Delphi Assessment: Expert Opinion, Forecasting and Group Process", R-1283-PR, April 1974. Brown, Thomas, "An Experiment in Probabilistic Forecasting", R-944-ARPA, 1972

4. Prasanna Chandra, “Projects: Planning, analysis, selection, financing implementation and review”

16 | P a g e