An essential aspect of managing any organization is planning for the future.

Indeed, the long-run

success of an organization is closely related to how well management is able to foresee the future and
develop appropriate strategies. Good judgment, intuition, and an awareness of the state of the
economy may give a manager a rough idea or "feeling" of what is likely to happen in the future.
However, it is often difficult to convert this feeling into a number that can be used as next quarter's
sales volume or next year's raw material cost per unit. The purpose of this chapter is to introduce
several forecasting methods.
Suppose we have been asked to provide quarterly forecasts of the sales volume for a particular
product during the coming 1-year period. Production schedules, raw material purchasing plans,
inventory policies, and sales quotas will all be affected by the quarterly forecasts that we provide.
Consequently, poor forecasts may result in increased costs for the firm. How should we go about
providing the quarterly sales volume forecasts?

We will certainly want to review the actual sales data for the product in past periods. Suppose we
have actual sales data for each quarter over the past 3 years. Using these historical data, we can
identify the general level of sales and determine whether there is any trend, such as an increase or
decrease in sales volume over time. A further review of the data might reveal a seasonal pattern, such
as peak sales occurring in the third quarter of each year and sales volume bottoming out during the
first quarter. By reviewing historical data over time, we can often develop a better understanding of
the pattern of past sales; often this can lead to better predictions of future sales for the product.

The historical sales data form a time series. A time series is a set of observations of a variable
measured at successive points in time or over successive periods of time. In this chapter we will
introduce several procedures for analyzing a time series. The objective of such analyses is to provide
good forecasts or predictions of future values of the time series.
Forecasting methods can be classified as quantitative or qualitative. Quantitative forecasting methods
can be used when (1) past information about the variable being forecast is available; (2) the
information can be quantified; and (3) a reasonable assumption is that the pattern of the past will
continue into the future. In such cases, a forecast can be developed using a time series method or a
causal method.

If the historical data are restricted to past values of the variable that we are trying to forecast, the
forecasting procedure is called a time series method. The objective of time series methods is to
discover a pattern in the historical data and then extrapolate this pattern into the future; the forecast
is based solely on past values of the variable that we are trying to forecast and/or on past forecast
errors. In this chapter we discuss three time series methods: smoothing (moving averages, weighted
moving averages, and exponen- tial smoothing), trend projection, and trend projection adjusted for
seasonal influence.

Causal forecasting methods are based on the assumption that the variable that we are trying to
forecast exhibits a cause-effect relationship with one or more other variables. In' this chapter we
discuss the use of regression analysis as a causal forecasting method. For instance, the sales volume
for many products is influenced by advertising expenditures, so regression analysis may be used to
develop an equation showing how these two variables are related. Then, once the advertising budget
has been set for the next period, we could substitute this value into the equation to develop a
prediction or forecast of the sales volume for that period. Note that if a time series method had been
used to develop the forecast, advertising expenditures would not even have been considered; that is,
a time series method would have based the forecast solely on past sales.

Qualitative methods generally involve the use of expert judgment to develop forecasts. For instance, a
panel of experts might develop a consensus forecast of the prime rate for a year from now. An
advantage of qualitative procedures is that they can be applied when the information on the variable
being forecast cannot be quantified and when historical data either are not applicable or available.
Figure 16.1 provides an overview of the types of forecasting methods.

using this flow, briefly discuss each bullet with the article as bases, 1. What is forecasting:
definition and importance in business decision-making 2. Types of forecasting methods:
qualitative, quantitative, and hybrid approaches 3. Qualitative forecasting methods: expert
opinion, market research, and Delphi method 4. Quantitative forecasting methods: time series
analysis, causal analysis, and regression analysis 5. Time series analysis: moving averages,
exponential smoothing, and trend analysis 6. Causal analysis: regression analysis, correlation
analysis, and econometric modeling 7. Accuracy measures for forecasting: mean absolute
deviation, mean squared error, and mean absolute percentage error 8. Factors that can affect
the accuracy of forecasts: data quality, seasonality, trend, and external factors such as
economic changes or unexpected events 9. Applications of forecasting: sales forecasting,
production forecasting, financial forecasting, and demand forecasting 10. Limitations of
forecasting: uncertainty and unpredictability, limitations of data and models, and the need for
ongoing evaluation and adjustment

diagram

Quantitative

Causal (Section 16.5)

Time series

Forecasting methods

Qualitative (Section 16.6)

Smoothing (Section 16.2)

Trend projection (Section 16.3)

Trend projection adjusted for seasonal influence (Section 16.4)

Report

Forecasting is crucial for businesses to make informed decisions about the future, as it
involves predicting outcomes based on historical data and other relevant information.
This process of estimating future values or events is essential for managing
organizations and helps in making decisions related to production schedules,
inventory policies, sales quotas, and raw material purchasing plans. Different types of
forecasting methods are available to businesses, and understanding their applications
is important for successful decision-making.

Qualitative

Qualitative forecasting is a type of forecasting method that involves analyzing

subjective information to make predictions about future outcomes. This approach is
useful when historical data is limited or unavailable, or when the future is highly
uncertain and difficult to predict using quantitative methods. Examples of qualitative
forecasting methods include expert opinion, market research, and the Delphi method.

Quantitative

Quantitative forecasting is a type of forecasting method that uses mathematical and

statistical models to analyze historical data and make predictions about future
outcomes. This approach is useful when there is a large amount of historical data
available, and when the future can be predicted based on patterns and trends
identified in the data.

Causal

Causal analysis is a type of forecasting method that looks at the relationship between
different variables to make predictions. This approach is useful when there is a cause-
and-effect relationship between variables, and when historical data can be used to
identify patterns and trends.

Regression analysis is one type of causal analysis that involves identifying the
relationship between a dependent variable (such as sales) and one or more
independent variables (such as advertising spend or price). This method can be used
to predict the impact of changes in independent variables on the dependent variable,
such as forecasting the effect of a price increase on sales.
Time series

Time series analysis is a type of forecasting method that involves analyzing historical
data over time to identify patterns and trends. This method can be used to make
short-term or long-term predictions, such as forecasting sales for the next month or
predicting demand for a product over the next year. Time series analysis can be used
for a variety of applications, such as forecasting stock prices or predicting traffic
patterns.

Smoothing

Smoothing is a technique used in time series analysis to reduce the impact of random
fluctuations or noise in the data, and to identify underlying trends and patterns. The
basic idea behind smoothing is to replace each data point with a weighted average of
nearby data points, in order to smooth out any random variations in the data.

For example, a company might use a moving average to forecast sales of a

particular product. By calculating the average sales for the past 12 months, the
company can smooth out any short-term fluctuations in sales and identify the overall
trend in sales over the past year. This can be used to make predictions about future
sales and adjust production and marketing strategies accordingly.

Trend projection

Trend projection is a type of time series analysis that involves identifying and
forecasting trends in the data, typically by fitting a mathematical model to the data.
This approach is useful when the data exhibits a clear and consistent trend over time.

For example, a company might use trend projection to forecast sales of a

particular product. By fitting a linear regression model to historical sales data, the
company can identify the overall trend in sales over time and make predictions about
future sales based on this trend. This can be used to adjust production and marketing
strategies to meet expected demand.

Trend projection adjusted for seasonal influence

Trend projection adjusted for seasonal influence is a type of time series analysis that
combines trend projection with a seasonal adjustment factor. This approach is useful
when the data exhibits a clear and consistent trend over time, as well as regular
seasonal fluctuations.

One example of trend projection adjusted for seasonal influence would be

predicting ice cream sales. Ice cream sales are likely to follow a seasonal pattern, with
higher sales during the summer months and lower sales during the rainy months. To
predict ice cream sales using trend projection adjusted for seasonal influence, we
would need to account for both the overall trend in sales over time and the seasonal
fluctuations in sales.

Accuracy measures for forecasting: To assess the accuracy of forecasts, businesses use
different measures, such as mean absolute deviation, mean squared error, and mean
absolute percentage error. These measures help to evaluate the performance of
different forecasting methods and models.

Factors that can affect the accuracy of forecasts: Various factors can affect the accuracy
of forecasts, such as the quality of data used, seasonality, trend, and external factors
such as unexpected events or changes in the economy. Businesses should take these
factors into account when using forecasting to inform decision-making.

Applications of forecasting: Forecasting has numerous applications in business, such as

sales forecasting, production forecasting, financial forecasting, and demand
forecasting. These applications help businesses to plan and make informed decisions
about resource allocation and investment.

Limitations of forecasting: It is important to note the limitations of forecasting, such as

uncertainty and unpredictability, limitations of data and models, and the need for
ongoing evaluation and adjustment. Businesses should be aware of these limitations
and take them into account when using forecasting to inform decision-making

1. Qualitative forecasting techniques: These are subjective techniques that rely on

expert opinion and judgment rather than statistical analysis. They can be useful
when historical data is limited or unreliable. Examples of qualitative techniques
include market research, surveys, and Delphi method.
2. Quantitative forecasting techniques: These are objective techniques that rely on
statistical analysis of historical data to make predictions. Examples of
quantitative techniques include time series analysis, regression analysis, and
econometric modeling.
3. Time series forecasting techniques: These techniques are used to make
predictions based on patterns and trends observed in historical data. Examples
include moving averages, exponential smoothing, and ARIMA models.
4. Causal forecasting techniques: These techniques are used to make predictions
based on causal relationships between variables. Examples include regression
analysis, input-output analysis, and simulation modeling.
5. Judgmental forecasting techniques: These techniques involve relying on the
judgment of experts or managers to make predictions. Examples include
scenario analysis, decision trees, and forecasting by analogy.
6. Ensemble forecasting techniques: These techniques involve combining the
forecasts of multiple models or experts to arrive at a more accurate prediction.
Examples include Bayesian model averaging, ensemble regression, and
combination forecasting.