ASSOCIATIVE FORECASTING TECHNIQUES:

SIMPLE LINEAR REGRESSION, COMMENTS,

AND CURVILINEAR AND MULTIPLE

ACCURACY AND CONTROL OF FORECASTS:

SUMMARIZING FORECAST ACCURACY

By: Ms. Myka Angela L. Rosales

 Associative Techniques
Rely on identification of related variables that can be used to predict values of the variable of
interest

Examples would be: 1) Sales of beef may be related to the price per pound charged for beef and the
prices of substitutes such as chicken, pork, and lamb; 2) Real estate prices are usually related to property
location and square footage and; 3) Crop yields are related to soil conditions and the amounts and timing
of water and fertilizer applications
Predictor Variables
Its effects are summarized by the development of an equation through associative techniques

These are variables used to predict values of the variable of interest.

Regression
Known as the primary method of analysis
A technique for fitting a line to a set of points
ASSOCIATIVE FORECASTING TECHNIQUES
A. SIMPLE LINEAR REGRESSION
The simplest and most widely used form of regression involves a linear relationship between
two variables.

Its object is to obtain an equation of a straight line

ASSOCIATIVE FORECASTING TECHNIQUES
ASSOCIATIVE FORECASTING TECHNIQUES
The coefficients a and b of the line are based on the following two equations:
ASSOCIATIVE FORECASTING TECHNIQUES

Example: Healthy Hamburgers has a chain

of 12 stores in Sydney. Sales figures and
profits for the stores are given in the
following table. Obtain a regression line for
the data and predict profit for a store
assuming sales of $10 million.
ASSOCIATIVE FORECASTING TECHNIQUES
Solution:

Plot the data and decide if the

linear model is reasonable.
ASSOCIATIVE FORECASTING TECHNIQUES
Solution:

Using the appropriate Excel

template on the text website,
obtain the regression
equation,
 ASSOCIATIVE FORECASTING TECHNIQUES
Indicators
Useful in the application of regression in forecasting

These are uncontrollable variables that tend to lead or precede changes in a variable of
interest

Careful identification and analysis of indicators may yield insight into possible future
demand in some situations
ASSOCIATIVE FORECASTING TECHNIQUES

These include: 1) Net change in inventories on hand and on order; 2) Interest rates for
commercial loans; 3) Industrial output; 4) Consumer price index (CPI); 5) The wholesale price
index; 6) Stock market prices; 7) Population shifts; 8) Local political climates; 9) Activities of
other firms (e.g., the opening of a shopping center may result in increased sales for nearby
businesses.)
ASSOCIATIVE FORECASTING TECHNIQUES

Three conditions are required for an indicator to be valid:

1. The relationship between movements of an indicator and movements of the variable should have
a logical explanation.
2. Movements of the indicator must precede movements of the dependent variable by enough time
so that the forecast isn’t outdated before it can be acted upon.
3. A fairly high correlation should exist between the two variables.
 ASSOCIATIVE FORECASTING TECHNIQUES
Correlation
Measures the strength and direction of relationship between two variables
It can range from -1.00 to +1.00
+1.00 – indicates that changes in one variable are always matched by the changes in the other.
-1.00 – indicates that increases in one variable are matched by decreases in the other.
Closes to zero – indicates little linear relationship between two variables
 ASSOCIATIVE FORECASTING TECHNIQUES
The correlation between two variables can be computed using the formula:

The square of the correlation coefficient, , provides a measure of the percentage of variability in the
values of that is “explained” by the independent variable. The possible values of range from 0 to 1.00
ASSOCIATIVE FORECASTING TECHNIQUES
Closer is to 1.00 – the greater the percentage of explained variation

High value of (0.80 or more) – indicates that the independent variable is a good
predictor of values of the dependent variable

Low value of (0.25 or less) – indicates a poor predictor

Value between 0.25 – 0.80 – indicates a moderate predictor

ASSOCIATIVE FORECASTING TECHNIQUES
B. COMMENTS ON THE USE OF LINEAR REGRESSION ANALYSIS
Assumptions on the use of simple regression analysis:

Variations around the line are random. If they are random, no patterns such as cycles or trends should be

apparent when the line and data are plotted.

Deviations around the line should be normally distributed. A concentration of values close to the line with a

small proportion of larger deviations supports the assumption of normality.

Predictions are being made only within the range of observed values.
ASSOCIATIVE FORECASTING TECHNIQUES
Observations to obtain the best results of the assumptions:

Always plot the data to verify that a linear relationship is appropriate

The data may be time-dependent. Check this by plotting the dependent variable versus
time; if patterns appear, use analysis of time series instead of regression, or use time as an
independent variable as part of a multiple regression analysis.

A small correlation may imply that other variables are important.

ASSOCIATIVE FORECASTING TECHNIQUES
Weaknesses of regression:

Simple linear regression applies only to linear relationships with one independent variable.

One needs a considerable amount of data to establish the relationship—in practice, 20 or

more observations

All observations are weighted equally

ASSOCIATIVE FORECASTING TECHNIQUES
Example:
Sales of new houses and three-month lagged unemployment are shown in the following table.
Determine if unemployment levels can be used to predict demand for new houses and, if so, derive
a predictive equation.
ASSOCIATIVE FORECASTING TECHNIQUES

Solution:

Plot the data to see if a linear model

seems reasonable. In this case, a linear
model seems appropriate for the range
of the data.
ASSOCIATIVE FORECASTING TECHNIQUES

Check the correlation coefficient to confirm that it is not close to zero using the
website template, and then obtain the regression equation:
ASSOCIATIVE FORECASTING TECHNIQUES
C. CURVILINEAR AND MULTIPLE REGRESSION
Reasons why simple regression analysis is inadequate to handle certain problems:
1. A linear model is inappropriate 2. more than one predictor variable is involved
Curvilinear Regression
Must be employed when nonlinear relationships are present
Model that involve more than one predictor

Require the use of multiple regression analysis through computers than manual
ACCURACY AND CONTROL OF FORECASTS
Accuracy and Control of Forecasts
A vital aspect of forecasting, so forecasters want to minimize forecast errors

Those involved in some forecasting applications:

Series of forecasts – used for weekly revenues

Single forecast – used for a one-time decision (e.g., the size of a power plant)

Periodic forecasts – it is important to monitor forecast errors to determine if the errors are
within reasonable bounds. If not, take corrective action.
ACCURACY AND CONTROL OF FORECASTS
Forecast Errors
The difference between the value that occurs and the value that was predicted for a given
time period
Hence, Error = Actual – Forecast
ACCURACY AND CONTROL OF FORECASTS
Positive Errors Negative Errors
Result when the forecast is too low When the forecast is too high

For example, if actual demand for a week is 100 units and forecast demand was 90 units, the forecast was
too low; the error is 100 - 90 = +10

Forecast errors influence decisions in two different ways:

1. In making a choice between various forecasting alternatives
2. In evaluating the success or failure of a technique in use
ACCURACY AND CONTROL OF FORECASTS

A. SUMMARIZING FORECAST ACCURACY

Forecast Accuracy

A significant factor when deciding among forecasting alternatives

Accuracy

Based on the historical error performance of a forecast

ACCURACY AND CONTROL OF FORECASTS
Three commonly used measures to summarize historical errors:
ACCURACY AND CONTROL OF FORECASTS
Example: Compute MAD, MSE, and MAPE for the following data, showing actual and forecasted
numbers of accounts serviced.
ACCURACY AND CONTROL OF FORECASTS
Solution: Using the figures shown in the table,
 OBJECTIVES
ACCOMPLISHED!!!
5. Briefly describe averaging techniques,
trends, and seasonal techniques, and
regression analysis, and solve typical 6. Describe two measures of
problems forecast accuracy
- Solve this typical problem relates to what have - What are the two measures of forecast
been discussed (regression analysis). accuracy?
Problem:
The owner of a small hardware store has noted a sales pattern for window locks that seems to parallel
the number of break-ins reported each week in the newspaper. The data are:

Plot the data to determine which type of equation, linear or nonlinear, is appropriate.
Obtain a regression equation for the data.
Estimate average sales when the number of break-ins is five.
Solution:
The graph supports a linear relationship
Obtain the regression coefficients using the
appropriate excel template. Simply replace the
existing data for x and y with you data.
Problem:
The manager of a large manufacturer of industrial pumps must choose between two alternative
forecasting techniques. Both techniques have been used to prepare forecasts for a six-month period. Using
MAD as a criterion, which technique has the better performance record?
Solution: Check that each forecast has an average error of approximately zero. (See computations that
follow.

Technique 1 is superior in this comparison because its MAD is smaller, although six observations would
generally be too few on which to base a realistic comparison.
Problem:
Given the demand data that follow, prepare a naive forecast for periods 2 through 10. Then determine
each forecast error, and use those values to obtain 2s control limits. If demand in the next two periods turns
out to be 125 and 130, can you conclude that the forecasts are in control?
Solution: For a naive forecast, each period’s demand becomes the forecast for the next period. Hence, the forecasts and errors
are:

The control limits are: 2(4.33) = ± 8.66. The forecast for period 11 was 124. Demand turned out to be 125, for an error
of 125 – 124 = +1. This is within the limits of ± 8.66. If the next demand is 130 and the naive forecast is 125 (based on
the period 11 demand of 125), the error is +5. Again, this is within the limits, so you cannot conclude the forecast is not
working properly. With more values – at least five or six – you could plot the errors to see whether you could detect
any pattern suggesting the presence of nonrandomness.
 ACCURACY AND CONTROL OF
FORECASTS: CONTROLLING THE
FORECAST

CHOOSING A FORECASTING
TECHNIQUE, USING FORECAST
INFORMATION, AND COMPUTERS

By: Ms. Micaella Kathleen L. Tolentino

ACCURACY AND CONTROL OF FORECASTS
B. CONTROLLING THE FORECAST
Possible sources of forecasting errors:
1. The model may be inadequate due to:
Omission of important variables
Change or shift in the variable that the model cannot deal with

Appearance of a new variable

ACCURACY AND CONTROL OF FORECASTS
2. Irregular variations may occur due to severe weather or other natural phenomenon, temporary
shortages or breakdowns, catastrophes, or similar events.

3. The forecasting technique may be used incorrectly, or the results misinterpreted.

4. There are always random variations in the data.

Randomness
inherent variation that remains in the data after all causes of variation have been
accounted for
ACCURACY AND CONTROL OF FORECASTS
Control Chart

Useful tool for detecting non randomness in errors

Visual tool for monitoring forecast errors

Errors
Plotted on a control chart in the order that they occur
Centerline

Represents an error of zero

ACCURACY AND CONTROL OF FORECASTS
UPPER AND LOWER CONTROL LIMITS

The two other lines below and above

the center

They represents the upper and lower

ends of the range of acceptable
variation for the errors
ACCURACY AND CONTROL OF FORECASTS
Things necessary for the
forecasts errors to be judged
“in control”:
All errors are within
control limits
No patterns are present
ACCURACY AND CONTROL OF FORECASTS
Compute the MSE:

The square root of MSE is used in practice as an estimate of standard deviation of the distribution of
errors.

Control charts are based on the assumptions when errors are random, they will be distributed according to a normal
distribution around a mean of zero. For a normal distribution, approximately 95.5% of the values can be expected to fall
within limits of _ , and approximately 99.7% of the values can be expected to fall within
____ of zero.
ACCURACY AND CONTROL OF FORECASTS
Formula to obtain UCL and LCL:

Where: z - number of standard deviation from the mean

Combining these 2 formulas, we obtain the following expression for the control limits:

Compute 2s control limits for forecast errors when the MSE is 2.0.
ACCURACY AND CONTROL OF FORECASTS
Tracking Signal

An older less informative technique that is sometimes employed to monitor

forecast errors
ratio of cumulative forecast error to the corresponding value of MAD to monitor
a forecast
Its intention is to detect any bias in errors over time (I.e., a tendency for a
sequence of errors to be positive or negative)
ACCURACY AND CONTROL OF FORECASTS
Values can be positive or negative. A value of zero would be ideal; limits of ______ are often used for a
range of acceptable values of the tracking signal.

If a. value outside the acceptable range occurs, that would be taken that there is bias in
the forecast, and that corrective action is needed

After an initial value of MAD had been determined, MAD can be updated using exponential
smoothing:
ACCURACY AND CONTROL OF FORECASTS
Example: Monthly attendance at financial planning seminars for the past 24 months, and forecasts and
errors for those months, are shown in the following table. Determine if the forecast is working using
these approaches:

A tracking signal, beginning with month 10, updating MAD with exponential smoothing. Use limits of

A control chart with 2s limits. Use data from the first eight months to develop the
control chart, then evaluate the remaining data with the control chart
ACCURACY AND CONTROL OF FORECASTS
ACCURACY AND CONTROL OF FORECASTS
Solution: 1. The sum of absolute errors through the 10 th
month is 58. Hence, the initial MAD is 58/10 = 5.8. The
subsequent MADs are updated using the formula MADnew.
The tracking signal is:
Cumulative error at that
month
Updated MAD at that
month

Because the tracking signal is within __ every month, there is no evidence of a problem
ACCURACY AND CONTROL OF FORECASTS
2. Make sure that the average error is approximately zero, because a large average would suggest a
biased forecast.

Compute for the standard

deviation: Determine 2s control limits:
ACCURACY AND CONTROL OF FORECASTS
I. Check that all errors are within the limits. (They
are.)
II. Plot the data, and check for nonrandom patterns. Note the strings of positive
and negative errors. This suggests nonroandomness. The tracking signal did not
reveal this.
A plot helps you to visualize the process and enables you to check for possible
patterns within the limits that suggest and improved forecast is possible.
ACCURACY AND CONTROL OF FORECASTS
A control chart focuses attention on deviations that lie outside predetermined limits. With either approach,
however, it is desirable to check for possible patterns in the errors, even if all errors are within the limits.

If nonrandomness is found, corrective action is needed. That will result in less variability in forecast errors,
and thus, in narrower control limits.

Illustrates the impact on control limits due to decreased error variability

ACCURACY AND CONTROL OF FORECASTS
Remember!!!

The control chart approach is generally superior to the tracking signal approach. A major weakness of the
tracking signal device approach is its use of cumulative errors: Individual errors can be obscured so that
large positive and negative values cancel each other.

Conversely, with control charts, every error is judged individually. Thus, it can be misleading to rely on a
tracking signal approach to monitor errors. In fact, the historical roots of the tracking signal approach date from
before the first use of computers in business.
ACCURACY AND CONTROL OF FORECASTS

At that time, it was much more difficult to compute standard deviations than to compute average
deviations; for that reason, the concept of a tracking signal was developed. Now computers and
calculators can easily provide standard deviation.

Nonetheless, the use of tracking signals has persisted, probably because users are unaware of
the superiority of the control chart approach.
 CHOOSING A FORECASTING TECHNIQUE
Two important factors when selecting a techniques:
Cost
How much money is budgeted for generating forecast?
Accuracy

What are the possible cost of errors?

What are the benefits that might accrue from an accurate forecast?
 CHOOSING A FORECASTING TECHNIQUE

Other factors to consider when selecting a technique:

1. Availability of historical data
2. Availability of computer software
3. Time needed to gather and analyze data and to prepare a forecast

Short Range Techniques

Moving averages and exponential smoothing
 CHOOSING A FORECASTING TECHNIQUE

Long Range Techniques

Trend equations and qualitative techniques

Long Range Techniques

Delphi method and executive opinion methods

 USING FORECAST INFORMATION
Two approaches to a forecast:

Reactive Approach
Views forecast as probable future demand, and a manager reacts to meet that demand.

Proactive Approach
Seeks to actively influence demand
Requires an explanatory model or a subjective assessment of the influence on demand
 USING FORECAST INFORMATION

Two forecasts a manager must make:

1. To predict what will happen under the
status quo
2. Based on a “what if” approach
USING FORECAST INFORMATION
 COMPUTERS IN FORECASTING
Computers play an important role in preparing forecasts based on quantitative data. Their use allows
managers to develop and revise forecast quickly, and without the burden of manual computations.

Software Packages

EXAMPLE : SPREADSHEET APPROACH = Excel templates at the Online Learning Center

Forecasts are the basis for many decisions. Clearly the more accurate an organization’s forecasts, the better prepared it
will be to take advantage of future opportunities and reduce potential risks.
 COMPUTERS IN FORECASTING

Better short term forecasts will:

Enhance profits
Lower inventory levels
Fewer shortages
Improve customer service levels
Enhance forecasting credibility throughout the organization
 OBJECTIVES
ACCOMPLISHED!!!
7. Describe two ways of evaluating 8. Identify the major factors to
and controlling forecasts consider when choosing a
forecasting technique
- What are the two ways of evaluating and - What are the major focus to consider
controlling forecasts? when choosing a forecasting technique?
 CASE STUDY

M&L Manufacturing
.......
Highline Financial
Services, Ltd.
Thanks!
Any questions?