**Critically discuss different methods of demand forecasting.**

__Methods Based on Judgment__

Unaided judgment METHOD.

Unaided judgment METHOD

It is common practice to ask experts what will happen. This is a good procedure to use when • experts are unbiased • large changes are unlikely • relationships are well understood by experts (e.g., demand goes up when prices go down) • experts possess privileged information • experts receive accurate and well-summarized feedback about their forecasts.

__Prediction markets METHOD__.

Prediction markets, also known as betting markets, information markets, and futures markets have a long history.

Some commercial organisations provide internet markets and software that to allow participants to predict.Consultants can also set up betting markets within firms to bet on such things as the sales growth of a new product. PREDICTIONS can produce accurate sales forecasts when used within companies. However, there are no empirical studies that compare forecasts from prediction markets and with those from traditional groups or from other methods.

__Delphi METHOD__.

The Delphi technique helps to capture the knowledge of diverse experts while avoiding the disadvantages of traditional group meetings. The latter include bullying and time-wasting. To forecast with Delphi the administrator should recruit between five and twenty suitable experts and poll them for their forecasts and reasons. The administrator then provides the experts with anonymous summary statistics on the forecasts, and experts’ reasons for their forecasts. The process is repeated until there is little change in forecasts between rounds – two or three rounds are usually sufficient. The Delphi forecast is the median or mode of the experts’ final forecasts. The forecasts from Delphi groups are substantially more accurate than forecasts from unaided judgement and traditional groups, and are somewhat more accurate than combined forecasts from unaided judgement.

__Structured analogies METHOD__.

The outcomes of similar situations from the past (analogies) may help a marketer to forecast the outcome of a new (target) situation. For example, the introduction of new products in the markets can provide analogies for the outcomes of the subsequent release of similar products in other countries. People often use analogies to make forecasts, but they do not do so in a structured manner. For example, they might search for an analogy that suits their prior beliefs or they might stop searching when they identify one analogy. The structured-analogies method uses a formal process to overcome biased and inefficient use of information from analogous situations. To use the structured analogies method, an administrator prepares a description of the target situation and selects experts who have knowledge of analogous situations; preferably direct experience. The experts identify and describe analogous situations, rate their similarity to the target situation, and match the outcomes of their analogies with potential outcomes in the target situation. The administrator then derives forecasts from the information the experts provided on their most similar analogies. Structured analogies are more accurate than unaided judgment in forecasting decisions .

__Game theory METHOD.__

is a way to obtain better forecasts in situations involving negotiations or other conflicts. BUT IT IS NOT A RELIABLE METHOD.

__Judgmental Decomposition METHOD__.

The basic idea behind judgemental decomposition is to divide the forecasting problem into parts that are easier to forecast than the whole. One then forecasts the parts individually, using methods appropriate to each part. Finally, the parts are combined to obtain a forecast. One approach is to break the problem down into multiplicative components. For example, to forecast sales for a brand, one can forecast industry sales volume, market share, and selling price per unit. Then reassemble the problem by multiplying the components together. Empirical results indicate that, in general, forecasts from decomposition are more accurate than those from a global approach . In particular, decomposition is more accurate where there is much uncertainty about the aggregate forecast and where large numbers (over one million) are involved.

__Expert systems METHOD__.

As the name implies, expert systems are structured representations of the rules experts use to make predictions or diagnoses. For example, ‘if local household incomes are in the bottom quartile, then do not supply premium brands’. The forecast is implicit in the foregoing conditional action statement: i.e., premium brands are unlikely to make an acceptable return in the locale. Rules are often created from protocols, whereby forecasters talk about what they are doing while making forecasts. Where empirical estimates of relationships from structured analysis such as econometric studies are available, expert systems should use that information. Expert opinion, conjoint analysis, and bootstrapping can also aid in the development of expert systems. Expert systems forecasting involves identifying forecasting rules used by experts and rules learned from empirical research. One should aim for simplicity and completeness in the resulting system, and the system should explain forecasts to users. Developing an expert system is expensive and so the method will only be of interest in situations where many forecasts of a similar kind are required. Expert systems are feasible where problems are sufficiently well-structured for rules to be identified. Expert systems forecasts are more accurate than those from unaided judgement.

__Simulated interaction METHOD__

Simulated interaction is a form of role playing for predicting decisions by people who are interacting with others. It is especially useful when the situation involves conflict. For example, one might wish to forecast how best to secure an exclusive distribution arrangement with a major supplier. To use simulated interaction, an administrator prepares a description of the target situation, describes the main protagonists’ roles, and provides a list of possible decisions. Role players adopt a role and read about the situation. They then improvise realistic interactions with the other role players until they reach a decision; for example to sign a trial one-year exclusive distribution agreement. The role players’ decisions are used to make the forecast. Forecasts from simulated interactions were substantially more accurate than can be obtained from unaided judgement. Simulated interaction can also help to maintain secrecy. Information on simulated interaction is available from conflictforecasting.com.

Intentions and expectations surveys METHOD.With intentions surveys, people are asked how they intend to behave in specified situations. In a similar manner, an expectations survey asks people how they expect to behave. Expectations differ from intentions because people realize that unintended things happen. For example, if you were asked whether you intended to visit the dentist in the next six months you might say no. However, you realize that a problem might arise that would necessitate such a visit, so your expectations would be that the event had a probability greater than zero.

Expectations and intentions can be obtained using probability scales . The scale should have descriptions such as 0 = ‘No chance, or almost no chance (1 in 100)’ to 10 = ‘Certain, or practically certain (99 in 100)’. To forecast demand using a survey of potential consumers, the administrator should prepare an accurate and comprehensive description of the product and conditions of sale. He should select a representative sample of the population of interest and develop questions to elicit expectations from respondents. Bias in responses should be assessed if possible and the data adjusted accordingly. The behaviour of the population is forecast by aggregating the survey responses.

__Conjoint analysis METHOD__.

By surveying consumers about their preferences for alternative product designs in a structured way, it is possible to infer how different features will influence demand. Potential customers might be presented with a series of perhaps 20 pairs of offerings. For example, various features of a personal digital assistant such as price, weight, battery life, screen clarity and memory could be varied substantially such that the features do not correlate with one another. The potential customer is thus forced to make trade-offs among various features by choosing one of each pair of offerings in a way that is representative of how they would choose in the marketplace. The resulting data can be analysed by regressing respondents’ choices against the product features. The method is based on sound principles, such as using experimental design and soliciting independent intentions from a sample of potential customers. Unfortunately however, there do not appear to be studies that compare conjoint-analysis forecasts with forecasts from other reasonable methods.

__Methods requiring quantitative data__

__Extrapolation METHOD__

Extrapolation methods use historical data on that which one wishes to forecast. Exponential smoothing is the most popular and cost effective of the statistical extrapolation methods. It implements the principle that recent data should be weighted more heavily and ‘smoothes’ out cyclical fluctuations to forecast the trend. To use exponential smoothing to extrapolate, the administrator should first clean and deseasonalise the data, and select reasonable smoothing factors. The administrator then calculates an average and trend from the data and uses these to derive a forecast Statistical extrapolations are cost effective when forecasts are needed for each of hundreds of inventory items. They are also useful where forecasters are biased or ignorant of the situation . Allow for seasonality when using quarterly, monthly, or daily data. Most firms do this . Seasonality adjustments led to substantial gains in accuracy in the large-scale study of time series .

__Quantitative analogies METHOD.__

Experts can identify situations that are analogous to a given situation. These can be used to extrapolate the outcome of a target situation. For example, to assess the loss in sales when the patent protection for a drug is removed, one might examine the historical pattern of sales for analogous drugs. To forecast using quantitative analogies, ask experts to identify situations that are analogous to the target situation and for which data are available. If the analogous data provides information about the future of the target situation, such as per capita ticket sales for a play that is touring from city to city, forecast by calculating averages. If not, construct one model using target situation data and another using analogous data. Combine the parameters of the models, and forecast with the combined model.

__Rule-based forecasting METHODS__

Rule-based forecasting (RBF) is a type of expert system that allows one to integrate managers’ knowledge about the domain with time-series data in a structured and inexpensive way. For example, in many cases a useful guideline is that trends should be extrapolated only when they agree with managers’ prior expectations. When the causal forces are contrary to the trend in the historical series, forecast errors tend to be large . Although such problems occur only in a small percentage of cases, their effects are serious. To apply RBF, one must first identify features of the series using statistical analysis, inspection, and domain knowledge (including causal forces). The rules are then used to adjust data, and to estimate short- and long-range models. RBF forecasts are a blend of the short- and long-range model forecasts. RBF is most useful when substantive domain knowledge is available, patterns are discernable in the series, trends are strong, and forecasts are needed for long horizons. Under such conditions, errors for rule-based forecasts are substantially less than those for combined forecasts . In cases where the conditions were not met, forecast accuracy is not harmed.

__Neural nets METHODS__

Neural networks are computer intensive methods that use decision processes analogous to those of the human brain. Like the brain, they have the capability of learning as patterns change and updating their parameter estimates. However, much data is needed in order to estimate neural network models and to reduce the risk of over-fitting the data .There is some evidence that neural network models can produce forecasts that are more accurate than those from other methods . While this is encouraging, our current advice is to avoid neural networks because the method ignores prior knowledge and because the results are difficult to understand.

__Data mining METHODS__

Data mining uses sophisticated statistical analyses to identify relationships. It is a popular approach. Data mining ignores theory and prior knowledge in a search for patterns. Despite ambitious claims and much research effort, we are not aware of evidence that data mining techniques provide benefits for forecasting.

__Causal models METHODS.__

Causal models are based on prior knowledge and theory. Time-series regression and cross-sectional regression are commonly used for estimating model parameters or coefficients. These models allow one to examine the effects of marketing activity, such as a change in price, as well as key aspects of the market, thus providing information for contingency planning. To develop causal models, one needs to select causal variables by using theory and prior knowledge. The key is to identify important variables, the direction of their effects, and any constraints. One should aim for a relatively simple model and use all available data to estimate it . Surprisingly, sophisticated statistical procedures have not led to more accurate forecasts. In fact, crude estimates are often sufficient to provide accurate forecasts when using cross-sectional data .

=======================================================

General principles •

Managers’ domain knowledge should be incorporated into forecasting methods.

• When making forecasts in highly uncertain situations, be conservative. For example, the trend should be dampened over the forecast horizon.

• Complex methods have not proven to be more accurate than relatively simple methods. Given their added cost and the reduced understanding among users, highly complex procedures cannot be justified.

• When possible, forecasting methods should use data on actual behaviour, rather than judgments or intentions, to predict behaviour.

• Methods that integrate judgmental and statistical data and procedures (e.g., rule-based forecasting) can improve forecast accuracy in many situations.

• Overconfidence occurs with quantitative and judgmental methods.

• When making forecasts in situations with high uncertainty, use more than one method and combine the forecasts, generally using simple averages.

Methods based on judgment

• When using judgment, rely on structured procedures such as Delphi, simulated interaction, structured analogies, and conjoint analysis.

• Simulated interaction is useful to predict the decisions in conflict situations, such as in negotiations. • In addition to seeking good feedback, forecasters should explicitly list all the things that might be wrong about their forecast. This will produce better calibrated prediction intervals.

Methods based on statistical data

• With the proliferation of data, causal models play an increasingly important role in forecasting market size, market share, and sales.

• Methods should be developed primarily on the basis of theory, not data.

Finally, efforts should be made to ensure forecasts are free of political considerations in a firm. To help with this, emphasis should be on gaining agreement about the forecasting methods. Also, for important forecasts, decisions on their use should be made before the forecasts are provided. Scenarios are helpful in guiding this process.

## 1 comment:

Who knows where to download XRumer 5.0 Palladium?

Help, please. All recommend this program to effectively advertise on the Internet, this is the best program!

Post a Comment