Module – 2

Forecasting: Steps in forecasting process, approaches to forecasting, forecasts based on

judgment and opinion, analysis.

INTRODUCTION

Forecasts are the basis for budgeting, planning capacity, sales, production and inventory,
personnel, purchasing, and more.

Forecasts play an important role in the planning process because they enable managers to
anticipate the future so they can plan accordingly.

Forecasts help managers by reducing some of the uncertainty, thereby enabling them to develop
more meaningful plans.

A forecast is a statement about the future value of a variable such as demand. That is, forecasts
are predictions about the future. The better those predictions, the more informed decisions can
be.

Forecasts Affect Decisions and Activities throughout an Organization

• Accounting

• Finance

• Human resources

• Marketing

• Management information systems (MIS)

• Other parts of an organization.

Here are some examples of uses of forecasts in business organizations –

Accounting: New product/process cost estimates, profit projections, cash management.

Finance: Equipment/equipment replacement needs, amount of funding/borrowing needs.

Human resources: Hiring activities, including recruitment, interviewing, and training;

Marketing: Pricing and promotion, business strategies.

MIS: New/revised information systems, Internet services.

Operations: Schedules, capacity planning, inventory planning, outsourcing.

Product/service design: Revision of current features, design of new products or services.

1
In most of these uses of forecasts, decisions in one area have consequences in other areas.

Therefore, it is very important for all affected areas to agree on a common forecast.

Accurate forecasts can help managers plan tactics (e.g., offer discounts, don’t offer discounts) to
match capacity with demand, thereby achieving high productivity.

There are two uses for forecasts. One is to help managers plan the system, and the other is to
help them plan the use of the system.

Planning the system generally involves long-range plans about the types of products and
services to offer, what facilities, where to locate, and so on.

Planning the use of the system refers to short-range and intermediate- range planning, which
involve tasks such as planning inventory and workforce levels, planning purchasing and
production, budgeting, and scheduling.

FEATURES COMMON TO ALL FORECASTS

1. Assumes a causal system approach

2. Forecasts are not perfect

3. Forecasts for groups tend to be more accurate than forecasts for individual items.

4. Forecast accuracy decreases as the time period covered by the forecast increases

 Forecasting techniques generally assume that the same underlying causal system that
existed in the past will continue to exist in the future.

 Forecasts are not perfect; actual results usually differ from predicted values; Allowances
should be made for forecast errors.

 Forecasts for groups of items tend to be more accurate than forecasts for individual items
because forecasting errors among items in a group usually have a canceling effect.

 Forecast accuracy decreases as the time period covered by the forecast increases.
Generally speaking, short-range forecasts must contend with fewer uncertainties than
longer-range forecasts, so they tend to be more accurate.

ELEMENTS OF A GOOD FORECAST

1. Should be timely.

2. Should be accurate.

3. Should be reliable.

4. Expressed in meaningful units.

2
5. Should be in writing.

6. Techniques used should be simple to understand and use.

7. Should be cost-effective.

 The forecast should be timely. Usually, a certain amount of time is needed to respond to
the information contained in a forecast.

 The forecast should be accurate, and the degree of accuracy should be stated.

 The forecast should be reliable; A technique that sometimes provides a good forecast and
sometimes a poor one will leave users with the uneasy feeling.

 The forecast should be expressed in meaningful units.

 The forecast should be in writing.

 The forecasting technique should be simple to understand and use.

 The forecast should be cost-effective: The benefits should outweigh the costs.

STEPS IN THE FORECASTING PROCESS

1. Determine the purpose of the forecast

2. Establish a time horizon

3. Obtain, clean, and analyze appropriate data

4. Select a forecasting technique

5. Make the forecast

6. Pass on the forecast results to the respective departments for further action

7. Monitor the forecast

 Determine the purpose of the forecast. How will it be used and when will it be needed?

 Establish a time horizon. The forecast must indicate a time interval, keeping in mind
that accuracy decreases as the time horizon increases.

 Obtain, clean, and analyze appropriate data. Once data is obtained, the data may need
to be “cleaned” to get rid of outliers and obviously incorrect data before analysis.

 Select a forecasting technique.

 Make the forecast. Do analysis using mathematical models.

 Pass on the forecast results to the respective departments for further action

3
  Monitor the forecast. A forecast has to be monitored to determine whether it is
performing in a satisfactory manner. If it is not, reexamine and prepare a revised forecast.

APPROACHES [TYPES] TO FORECASTING

1. Qualitative Approach or Opinion and Judgmental Methods

2. Quantitative Approach or Time – Series Methods

Each type has different uses so it's important to pick the one that that will help you meet your
goals. And understanding all the techniques available will help you select the one that will yield
the most useful data for your company.

There are three general approaches to forecasting:

1. Judgmental Forecasts

2. Time-series Forecasts

3. Associative Forecasts

Judgmental forecasts - are inputs obtained from various sources, such as consumer surveys, the
sales staff, managers and executives, and panels of experts.

Time-series forecasts - Forecasts that project patterns identified in recent time-series

observations.

Associative model - Forecasting technique that uses explanatory variables to predict future
demand.

FORECASTING VARIABLES OR FACTORS AFFECTING FORECASTING

In any organization, generally speaking, forecasting activities are a function of the

1. Type of forecast

2. Time horizon being forecast

3. Data base available

4. Methodology of forecast

 Type of forecast: Any forecast must first establish the variables which are to be forecast.
Variables could be either controllable or non-controllable variables are those which can
be reasonably controlled by the management such as advertisement, budget, inventory
levels etc,. Whereas uncontrollable variables are those which are not in the hands of
management such as product demand, competition, raw material cost etc.

4
  Time horizon: Forecast can be made for different time periods and can be classified as
short term, long term and intermediate forecasts. Forecast technique can change with
changes in time horizons.
 Data base – Quantitative and Qualitative: Forecast technique become more
dependable if they use quantitative data rather than qualitative data. For example
information such as price of the product, previous demands, stock market index, etc are
more reliable than information such as brand image climate condition, etc.
 Forecasting Methodology: There are many forcasting methods available to business
organizations as listed in the succeeding paragraphs. There are both subjective and
objective methods. There is also both simple as well as complicated analysis. Forecast
methods tend to become complex as the amount of uncertainty about future events
increases.

CLASSIFICATION OF FORECASTING METHODS

I. Opinion and Judgmental Methods or Qualitative Approach

1) Opinion survey method

2) Market Trails method

3) Delphi Technique

4) Nominal Group Technique

II. Time – Series Methods or Quantitative Approach

1) Simple Average

2) Simple Moving Average

3) Weighted Moving Average

4) Simple Exponential Smoothing

5) Exponential Smoothing with Trend Adjustment.

III. Least squares or Regression Methods

I. OPINION AND JUDGMENTAL METHODS

1. Opinion survey or Market Research method: It is a relatively simple and practical method
for forecasting demands especially for new products. In this method, opinions are collected from
the prospective buyers regarding why they buy a particular product, what they expect from the
product and so on. The sampling technique is used to survey the customers. From the

5
information gathered it is possible to forecast how the targeted population responds to the
product.

2. Market Trails method: When a product concept is entirely new to the customer or market, it
is very difficult to anticipate the acceptability of the product. In such cases, a trial – run of the
product in the market is suggested. Such a trial is like a controlled experiment in which the
market area and the method of presentation are carefully selected and controlled.

3. Delphi Technique: It is a subjective method relying on the opinion of a few experts. This
method is designed in such a way as the minimize bias and error of judgment when compared to
other expert-opinion methods.

Advantages: a) The experts do not meet each other. Therefore there will not be any conflicts
between them.

b) There is no domination of one man over others

Disadvantages: a) It is a tedious and time-consuming method

b) Deadlocks are difficult to resolve.

4. Nominal Group Technique: It comprises of a panel of experts but here they work together in
a meeting to arrive at a consensus through discussion. The success of Nominal Group Process
lies in clearly identifying questions, allowing creativity & innovation, encouraging discussion
and ultimately arriving at a consensus.

II. TIME – SERIES METHODS

A time series is a time-ordered sequence of observations taken at regular intervals (e.g., hourly,
daily, weekly, monthly, quarterly, annually).

The data may be measurements of demand, earnings, profits, shipments, productivity, or the
consumer price index.

Time-series data are made on the assumption that future values of the series can be estimated
from past values.

No attempt is made to identify variables that influence the data

These methods are widely used, often with quite satisfactory results.

PATTERNS [COMPONENTS] IN TIME-SERIES

The most general type of time series is influenced by all four components, a
stable pattern consisting of trend, cyclical, and seasonal components observed in the presence of
the random error component

1. Trend refers to a long-term upward or downward movement in the data.

6
2. Seasonality refers to short-term, fairly regular variations generally related to factors such as
the calendar or time of day. Restaurants, supermarkets, and theaters experience weekly and even
daily “seasonal” variations.

3. Cycles are wavelike variations of more than one year’s duration. These are often related to a
variety of economic, political, and even agricultural conditions.

4. Irregular variations are due to unusual circumstances such as severe weather conditions,
strikes, or a major change in a product or service.

5. Random variations are residual variations that remain after all other behaviors have been
accounted for.

1) Simple Average(Moving Average)

A simple average is the average of the demands occurring in all previous periods. The demands
for all periods are equally weighted.

SA = Sum of demands for all periods / Number of periods

7
2) Simple Moving Average

This is nothing but average of demands occurring is a fixed number of recent periods. This is
known as moving average and is forecast as the future demand.

Where n= the choosen number of periods

t= 1 is the oldest period in the n – period average

i = n is the most recent period Di = the demand in the ith period

8
3) Weighted Moving Average

A weighted average is similar to a moving average, except that it assigns more weight to the
most recent values in a time series.

For instance, the most recent value might be assigned a weight of .40, the next most recent value
a weight of .30, the next after that a weight of .20, and the next after that a weight of .10.

Note that the weights must sum to 1.00, and that the heaviest weights are assigned to the most
recent values.

Worked Examples

2. The past data for the sales of wet grinders of a particular company in an area is shown below.

Month Sales

Jan 2001 585

Feb 2001 610

9
 March 2001 675

April 2001 750

May 2001 860

June 2001 970

Forecast the demand for the month of July 2001 using

a) Simple Average for all previous months

b) A- three-month moving average

c) A three-month moving average where the weights are 0.5 for the latest month, 0.3 and
0.2 for the months previous to that respectively.

Solution: a) Simple Average= Sum of demands for all periods / Number of periods

SA = 585 + 610 + 675 + 750 + 860 + 970 / 6

SA = 741.667

b) 3-month Moving Average (MA) = Sum of demands for periods / Choosen number of periods

MA = 750 + 860 + 970 / 3

MA = 860

Therefore for July 2001 using 3 month Moving Average = 860 Units

c) 3 month weighted moving average where weights are June=0.5 May=0.3 April=0.2

WMA = 0.5 x 970 + 0.3 x 860 + 0.2 x 750 = 893 Units

Since different methods give different forecasts, it is obvious that a certain method is selected
based on its performance as a forecast.

3. Compute Four-month Average and weighted moving average for the data given below.
Assume the weight for the most recent period is 3 times as that of the previous two periods.

Month Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec

Sales in 250 210 223 270 245 261 212 212 246 252 261 224
100’s

Solution: i) Four Month Moving Average (MA) = Sum of demands for periods / Choosen
number of periods

MA = (246 + 252 + 261 + 224) x 100 / 4 = 24575 Units

10
ii) Four Month Weighted Moving Average

If the weight for December has to three times the weight of October or November then by trial
and error, weight for

Dec = 0.54 Nov= 0.18 Oct=0.18 Sept=0.10 Total= 1.00

Therefore 4-month weighted moving average

WMA= 100(0.54 x 224 + 0.18 x 261 + 0.18 x 252 + 0.10 x 246)

WMA= 100 x 237.9

WMA= 23,790 Units

(Note: The above problem can have other answers as well. The weightage for different months
can be found from the relation [(5/3) Dec + Sept=1] in such a way that weight of Dec. is less
than 0.6, but on condition that September’s weight is the least. One more e.g., of giving
weightage is, Dec=0.51, Oct and Nov=0.17, Sept=0.15 and so on)

4) Simple Exponential Smoothing

In this type each new forecast is based on the previous forecast plus a percentage of the
difference between that forecast and the actual value of the series at that point.

Next forecast = Previous forecast + α (Actual - Previous forecast)

Ft = Ft-1 + α (Dt-1 – Ft-1)

Ft = Forecast for the time period‘t’

Ft-1 = Forecast for the time period‘t-1’ (Previous forecast)

α = Smoothing constant (0 <_ α <_1)

Dt-1 = Actual demand for the time period‘t-1’

For example,

Suppose the previous forecast was 42 units,

Actual demand was 40 units,

and α= 0.10.

The new forecast would be computed as follows:

11
Worked Problems

1. The demand for disposable plastic tubing in a hospital for September was 300 units and
for October were 350 units. Using 200 units as the September forecast and a smoothing
coefficient of 0.7 to weight recent demand more heavily, forecast the demand in
November.

Month Sept. Oct Nov.

Demand 300 350 ?

α = 0.7 Fsept = 200

Ft = Ft-1 + α (Dt-1 – Ft-1)

t= November then t-1 = October

Dt-1 =Demand for October = 350

Ft-1 = Forecast for September = 200

Ft = Forecast for October

Ft = Ft-1 + α (Dt-1 – Ft-1)

∴ Foct= 200 + 0.7(300-200)

Foct=270

Fnov = Foct + α (Doct – Foct) = 270 + 0.7 (350-270)

Fnov = 326 Units – This is the forecast for November

2. A firm uses simple exponential smoothing with α = 0.1 to forecast sales. The forecast for week
ending Feb 1 was 500 units whereas actual demand out to be 450 units.

i) Forecast the demand for week ending Feb.8

ii) Assume the actual demand during the week ending Feb 8 turned out to be 505 units.
Forecast the demand for week ending Feb 15. Continue the forecasting through March
15, assuming that the subsequent demands were actually 516, 488,467,554 and 510 units.

Ans, Let t= week ending Feb8

Then t-1= Previous week (i.e., week ending Feb 1)

Given α =0.1 Ft-1 = 500 Dt-1 =450

Ft = Ft-1 + α (Dt-1 – Ft-1)

12
∴ Ft = 500 + 0.1(450-500)

Ft=495 Units

Week – ending Old forecast Actual demand New forecast Ft = Ft-1 +

Dt-1
t-1 Ft-1 α (Dt-1 – Ft-1)

Feb 1 500 450 Ft = 500 +0.1(450 -500)=495

Feb 8 495 505 496

Feb 15 496 516 498

Feb 22 498 488 497

Mar 1 497 467 494

Mar 8 494 554 500

Mar 15 500 510 501

It is observed from the above solution table that every forecast becomes the old forecast for the
succeeding new forecast.

3. Using SES technique, determine the forecast for period 2 through 12 for which the actual
figures are given below.

Period 1 2 3 4 5 6 7 8 9 10 11 12

Actual Demand 200 211 190 198 210 230 195 200 215 198 200 212

Assume that the first period forecast is equal to actual demand in that period given α = 0.2. Also
graphically compare your forecast demand with actual demand.

Solution: In the following table, the new forecasts are stated to be calculated from period 2
onwards. Therefore period 1 becomes t-1 in the first row and so on it continues.

13
Period t- Old forecast Actual demand New forecast Ft = Ft-1 + α (Dt-1 – Ft-1)
1 Ft-1
Dt-1

1 200 200 F2= 200 + 0.2(200-200)=200

2 200 211 F3= 200 + 0.2(211-200)=202.20

3 202.20 190 F4= 202.20 + 0.2(190-202.20)=199.76

4 199.76 198 F5= 199.76 + 0.2(198-199.76)=199.40

5 199.40 210 F6= 199.40 + 0.2(210-199.40)=201.51

6 201.51 230 F7= 201.51 + 0.2(230-201.51)=207.20

7 207.20 195 F8= 207.20 + 0.2(195-207.20)=204.76

8 204.76 200 F9= 204.76 + 0.2(200-204.76)=203.80

9 203.80 215 F10= 203.80 + 0.2(215-203.80)=206.03

10 206.03 198 F11= 206.03 + 0.2(198-206.03)=204.42

11 204.42 200 F13= 204.42 + 0.2(230-204.42)=203.52

12 203.53 212

14
5) Exponential Smoothing with Trend Adjustment.

When the exponential smoothing technique takes into account seasonal effects along with regular
trend forecasts, we get exponential smoothing with trend adjustment or Adjusted exponential
smoothing.

Adjusted exponential smoothing models actually project into the future (for example, to time
period t+1) by adding a trend correction increment Tt, to the current – period smoothed average,
Ft.

∴ Trend adjustment Forecast = New Forecast + Trend Correction

Fta= Ft + Tt

Ft = Ft-1 + α (Dt-1 – Ft-1)

Tt = β(Ft- Ft-1) + (1 – β) Tt-1

Where Tt-1 = Smoothed trend for previous period

β = Smoothing constant for trend adjustment

Ft-1 = Forecast of previous period

Worked Problems

1. A manufacturer of printed circuit boards uses exponential smoothing with trend to

forecast monthly demand of its product. At the end of December the company wishes to
forecast sales for January. The estimate of trend through November has been 200
additional boards sold per month. Average sales have been around 1000 units per month.
The demand for December was 1100 units. The company uses α = 0.2 β = 0.1. Make a
forecast for the month of January.

Given

Trend for November, Tnov = 200 units

Forecast for every month, F = 1000 units

Demand for December, Ddec=1100 units

α = 0.2

β = 0.1

Fjan =?

Trend adjustment forecast, Fta = Ft + Tt

Or Fjan (ta) = Fjan + Tjan

15
Fjan = Fdec + α (Ddec – Fdec)

= 1000 + 0.2 (1100 -1000)

= 1020 units

Tt = β (Ft- Ft-1) + (1 – β) Tt-1

Tjan = β(Fjan- Fdec) + (1 – β) Tdec

We do not know Tdec, So we have to find it first.

Tdec= 0.1 ( 1000 – 1000) + (1 + 0.1) 200 Tdec= 180

Therefore Tjan= 180 + 0.1(1000 – 1000) Tjan= 180

Fjan (trend) = 1020 -180

Fjan (trend) = 1200 units.

Therefore the trend adjusted forecast for the month of January is 1200 units.

III. LEAST SQUARES OR REGRESSION METHODS

Least squares are a widely used mathematical method of obtaining line of best fit between the
dependent variable (usually demand) and an independent variable. This method is called least
squares method since the sum of the square of the deviations of the various points from the line
of best fit is minimum or least. It gives the equation of the line for which the sum of the squares
of vertical distances between the actual values and the line values are at minimum.

The least squares concept is also used for Regression and Correlation. Analysis between any set
of dependent and independent variables.

In Least squares or Regression Analysis, the relationship between the dependent variables Y and
some independent variable X can be represented by a straight line.

y=a+bX

Where b = Slope of straight line a = Y-intercept of the straight line

16
 Worked Problems

1. With the help of Least squares Method, develop a linear trend equation for the data
shown in the table.

i) Compute the constants a and b in the regression equation

ii) Forecast a trend value for the year 2002 and 2008.

Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Shipme 2 3 6 10 8 7 12 14 14 18 19
nts
(Tons)

Year X Year Coded Y Shipments(tons) XY X2

1991 -5 2 -10 25

1992 -4 3 -12 16

1993 -3 6 -18 9

1994 -2 10 -20 4

1995 -1 8 -8 1

1996 0 7 0 0

1997 1 12 12 1

1998 2 14 28 4

1999 3 14 42 9

2000 4 18 72 16

2001 5 19 95 25

n=11 ∑X=0 ∑Y=113 ∑XY=181 ∑X2=110

i) To find constants a and b

The straight line equation is Y = a + b X

17
Y = demand (shipment)

a= Y-intercept

b=Slope

X= Time period

Where a= ∑Y / n = 113/ 11 = 10.3

b= ∑XY / ∑X2 = 181/110 = 1.6

Y = 10.3 + 1.6 X

ii) To forecast for the years 2002 and 2008

It is observed from the table that if year 2001 is coded as +5, year 2002 would be +6 and year
2008 would be +12.

∴ For year 2002 Put X=6

Y = 10.3 + 1.6 x 6

Y = 19.9

∴ Forecast for the year 2002 = 19.9 tons of shipment

For year 2008 Put X=12

Y = 10.3 + 1.6 x 12

Y = 29.5

∴ Forecast for the year 2008 = 29.5 tons of shipment

2. The table below gives the sales record of a firm. Using Regression Analysis Forecast the sales
in the month of January and February next year.

Month Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec

Sales in 90 11 99 89 87 84 104 102 95 114 103 113

(units)

18
 Month X Month Coded Y Sales(units) X2 XY

January -5.5 90 30.25 -495

February -4.5 111 20.25 -499.5

March -3.5 99 12.25 -346.5

April -2.5 89 6.25 -222.5

May -1.5 87 2.25 -130.5

June -0.5 84 0.25 -42

July 0.5 104 0.25 52

August 1.5 102 0.25 153

September 2.5 95 6.25 237.5

October 3.5 114 12.25 399

November 4.5 103 20.25 463.5

December 5.5 113 30.25 621.5

n=12 ∑X=0 ∑Y=1191 ∑X2=143 ∑XY=190.5

[Note: In the above table, the values of X can also be coded as -11.5,-9.5,-7.5,-5.5,-3.5,-
1.5,+1.5,3.5,5.5,7.5,9.5,11.5. in which case also ∑X=0, but the difference between each month
coded is 2]

To find constants a and b

a= ∑Y / n = 1191/ 12 = 99.25

b= ∑XY / ∑X2 = 190.5/143 = 1.332

∴Y=a+bX

Y = 99.25 + 1.332 x X

Forecast for the month of January and February Next year

From the table, if the value of X for December is +5.5, the value of X for January would be +6.5
and that of February would be +7.5.

∴ For Jan, put X=6.5

Y= 99.25 + 1.332 x 6.5

19
Y= 107.9 or Y=108 units is the Forecast

∴ For Feb, put X=7.5

Y= 99.25 + 1.332 x 7.5

Y= 109.24 or Y=110 units is the Forecast

Case (ii) : When any quantity(other than time) is the independent variable

Normally the demand of any product would vary with time but in reality it depends on a variety
of factors like quality of the product, effectiveness of sales force, advertisement strategies and
budgets, distribution efficiencies, and so on. In such a case we consider demand to be dependent
on a quantity other than time. The procedure followed is the same as the previous case, but only
the formulae used to calculate the constants a and b are different. In case (ii) also a straight line is
fit whose equation is,

Y= a + b X

a=y – intercept

b= slope of straight line

Where a = [(∑X2 .∑Y) – (∑X. ∑XY)] / (n ∑X2) – (∑X2)

And b= [(n∑XY) – (∑X ∑Y)] / [(n ∑X2) – (∑X2 )]

Worked Problem:

1. A manufacturer of children’s cycle believes that the demand for the cycles is correlated to
the birth of babies in the area during the previous year.

The data given below shows this relationship.

Year No. of births in the previous year Cycles sold during the year

1 40,000 3,000

2 48,000 3,200

3 66,000 4,000

4 78,000 5,200

5 92,000 7,900

20
 6 1,05,000 7,900

7 1,25,000 9,000

8 1,40,000 10,000

Compute the probable sales of cycles in the 9th year given the no. of births in the previous year as
1, 66,000.

Regression analysis

Year No. of births in previous year Cycles sold this year Y X2 x 103 XY x 103
X x 103 x 103

1 40 30 1600 1200

2 48 32 2304 1536

3 66 37 4356 2442

4 78 40 6084 3120

5 92 52 8464 4784

6 105 79 11025 8295

7 125 90 15625 11250

8 140 100 19600 14000

n=8 ∑X=694 ∑Y=460 ∑X2=69058 ∑XY=46627

i) To find co-efficient’s ‘a’ and ‘b’ in the regression equation,

Y= a + b X

a = [(∑X2 .∑Y) – (∑X. ∑XY)] / (n ∑X2) – (∑X2)

a= (69058x4600) – (694 X 46627) / (8 X 69058) – (694)2

a = -8.36

b= [(n∑XY) – (∑X ∑Y) ] / [(n ∑X2 ) – (∑X2 )]

b=(8 x 46627) – (694x 460) / (8 x 69058) – (694)2

21
b=0.76

Y= a + b X

Y= -8.36 + 0.76X

If X is considered in multiples of 103, answer Y will be in multiples of 102, as defined in the

initial table.

If No. of births is X = 166000

Or X= 166 x 102

No. of cycles sold, Y =?

∴ Y = -8.36 + 0.76 x 166

Y = 117.8 or Y= 117.8 x 102

Y = 11780 cycles

This is corresponding expected sales of cycles in the 9th year when no. of births touches 166000.

CORRELATION CO-EFFICIENT

Regression Analysis basically tries to express the relationship between two variables in the form
of a straight line. The extent to which the two variables are related to each other is explained by
correlation analysis. In other words, correlation is a means of expressing the degree of
relationship between two or more variables.

A single figure which expresses the degree and direction of correlation is called co-efficient of
correlation®. The correlation coefficient(r) is a number between -1 and +1 and is designated as
positive if Y increases with increase in X and negative if Y decreases with increases in X. If r=0,
this indicates the lack of relationship between the two variable. Fig. illustrates the meaning of the
linear correlation coefficient.

22
Worked Problem:

1. The following data table gives the five months of average monthly temperatures and
corresponding monthly attendance.

Month 1 2 3 4 5

Average temperature 24 41 32 30 38

Resort 43 31 39 38 35
attendance(‘000s)

Compute linear regression equation of the relationship between the two if next month’s average
temperature is forecast to be 45 degrees.

i) Use linear regression equation to develop a forecast.

ii) Compute a correlation coefficient for the above data and determine the strength of the linear
relationship between temperature and attendance. How good predictor is temperature for
attendance?

Months Average Temp(ºC) Resort X2 Y2 XY

X Attendance

1 24 48 x 103 576 1849 x 106 1032 x 103

2 41 31 x 103 1681 961 x 106 1271 x 103

3 32 39 x 103 1024 1521 x 106 1248 x 103

23
 4 30 38 x 103 900 1444 x 106 1140 x 103

5 38 35 x 103 1444 1225 x 106 1330 x 103

n=5 ∑X=165 ∑Y=186 x 103 ∑X2 ∑Y2 ∑XY

=5625 =7000 x =6021 x

106 103

CONTROL OF FORECAST (or Forecast Errors)

Forecasts errors tend to be inaccurate, and you need to find out how (in) accurate
your forecasting model is. Forecasting error is the difference between the forecast and actual
values. Forecasts are inaccurate for many reasons. Correctly identifying variables has an impact
on your forecast.

24
25
 WORKED EXAMLEMS FROM QUESTION PAPERS

1. A company adopts method of least squares to develop a linear trend equation for the data
as shown in the table below:

Year(x) 1 2 3 4 5 6 7 8 9 10 11

Shipment in
2 3 5 10 8 7 12 14 14 18 19
tones(y)

Calculate the trend forecast for the year 12 and 20. (10 Marks) Dec.2019/Jan.2020

26
2. A firm use simple exponential smoothing with α =0.1 to forecast demand. The forecast for the
week of February 1 was 500 units, whereas actual demand turned out to be 450 units.

i) Forecast the demand for the week of February 8

ii) Assume that the actual demand during the week of February 8 turned out to be 505 units.
Forecast the demand for the week of February 15. Continue on forcasting through March 15,
assuming the subsequent demands were actually 516, 488, 467,554 and 510 units.
(07 Marks) Aug/Sept.2020

27
3. The general manager of a building materials production plant feels that the demand for
plasterboard shipments may be related to the number of construction permits issued in the
country during the previous quarter. The manager has collected the data shown in the
accompanying table. Determine the regression line. Find the forecast for plasterboard shipments
when the number of construction permits is 30.

28