Forecasting with examples

1. (Dr. M K Manik)
HOD.mechccetb@gmail.com

2. Sales forecasting
Sales forecasting is the process of estimating
future sales
Accurate sales forecasts help companies to make
business decisions
They calculate short-term and long-term goal of
company performance
Companies can base their forecasts on
past sales data
 Industry can do wide comparisons, and economic
trends
Methods of Sale forecasting
 Every manufacturer makes an estimation of the
sales for future

3. What is Economic Indicator?
An economic indicator is a piece
of economic data,
Usually of macroeconomic scale, that is used
by analysts to interpret current or future
investment possibilities
It is used to judge the overall health of
an economy.
Factors on which Economic Indicator
depends?
GDP ( Gross domestic product )
PMI (Purchasing Manager Index)

4. Purchasing Managers
 Index The Purchasing Managers' Index (PMI) is
an indicator of economic health for
manufacturing and service sectors
The purpose of the PMI is to provide information
about current business conditions to company
decision makers, analysts and purchasing
managers.
Consumer price index
A measure of changes in the purchasing-power
of a currency and the rate of inflation.
The consumer price index expresses the current
prices of a basket of goods and services in terms
of the prices during the same period in a previous

5. These Will Be The Top 15
Richest Countries In 2050
2 China - $25.33 trillion. The
richest country in the world in
2050 is predicted to be China
3 United States - $22.27 trillion
4 India - $8.17 trillion
5 Japan - $6.43 trillion
6 Germany - $3.71 trillion
7 United Kingdom - $3.58
trillion
8 Brazil - $2.96 trillion
Below are the top
10 most
developed states
in India 2018.
Tamil Nadu.
Kerala.
Maharashtra.
Karnataka.
Andhra Pradesh.
Rajasthan.
Uttar Pradesh.
Haryana.

6. Which is the poor state in India
Chhattisgarh,
Manipur,
Odisha
Madhya Pradesh,
Jharkhand,
Bihar
And Assam
figure among the poorest states where over 40 per cent of people are
below poverty line, according to the C Rangarajan panel
What do you mean by GDP
A. The GDP or gross domestic product of a country provides
a measure of the monetary value of the goods and
services that country produces in a specific year.
B. This is an important statistic that indicates whether an
economy is growing or contracting.

9. Forecast Topic: Moving Average Methods
One of the easiest, most common type of forecasting
techniques is that of the moving average
Moving average methods come in handy if several
consecutive periods of data is available
In this forecasting method next period’s sales are only
predicted
Often based on the past few months of sales the prediction
is dine for coming month’s sales
 However, moving average methods can have serious
forecasting errors if applied carelessly.

11. Problem-1 Demand for an item is observed for 15 months and data are given
below
Calculate i) 3 months and ii) 4 months moving average. and what is the forecast
for the month of 16. for each case.

12. Limitations of Moving Average Methods
Moving averages are considered a “smoothing”
forecast technique
 Because you’re taking an average over time
You are softening (or smoothing out) the effects
of irregular occurrences within the data
 As a result, the effects of seasonality, business
cycles, and other random events can dramatically
increase forecast error
Take a look at a full year’s worth of data, and
compare a 3-period moving average and a 5-period

13. Month Actual 3-Mo. Forecast Deviation
Absolute
Deviation
January 135 127 (8) 8
February 134 135 1 1
March 125 128 3 3
Rectification on moving average Method

14. Moving Averages: Recap
When using moving averages for forecasting,
remember:
Moving averages can be simple or weighted
The number of periods you use for your average,
and any weights you assign to each are strictly
arbitrary
Moving averages smooth out irregular patterns in
time series data; the larger the number of periods
used for each data point, the greater the smoothing
effect
Because of smoothing, forecasting next month’s
sales based on the most recent few month’s sales
can result in large deviations because of
seasonality, cyclical, and irregular patterns in the

17. Exponential Smoothing average Method
In this method the forecasting could be done based on the
calculation.
Here am Mathematical formulation such as Ft+1 = α At +
(1+α) Ft
Where Ft+1 = Fore cast for the next period with
respect to t ;
At = actual sales/demand for period of t.
α= Smoothing constant, 0 ≥ α ≥1; any value When
no value of α is given take any value between 0 to 1, Here I have taken α = 0.3
Ft= Forecast for time t .
Week Sales Forecast Ft+1 = α At + (1- α) Ft
1 39
F2 = α At + (1+α) At =0.3*39+(1-
0.3)39=39
2 44 Ft+1 = α At + (1+α) Ft =0.3*39+(1-
0.3)39=39
3 40
Ft+1 = α At + (1+α) l
4 45

18. Statistical Sales forecast show in graph

19. Problem1: Export an Item as shown in the following forecasting method, Fit a
straight-line by forecasting in the year of 2016 and 2017.
200
7
200
8
200
9
201
0
201
1
201
2
201
3
201
4
201
5
201
6
201
7
13 20 20 28 30 32 33 38 43 ? ?We know equation of straight line is Y = C + m X
We know normal equation for curve fittings are  ΣY = n c +m ΣX ---------
(1)
Here independent variable year as X and sales as Y.  ΣXY = c ΣX +m Σ --(2)Year
X
Demand
Y
ΣXY
0 13 0 0
1 20 20 1
2 20 40 4
3 28 84 9
4 30 120 16
5 32 160 25
6 33 198 36
7 38 266 49
8 43 344 64
ΣX=36 ΣY=257 1232 204
N= No. of terms is 9 (0,1,2,3, ---7,8 total 9 ).
Here in this problem 20070; 20081;
…20147;20158
Or N= No. of times of Independent vales[ value
of (x)]
Now you put the value of ΣX, ΣY, ΣXY
And Σ in the above two equations and find
out the value of coefficient C & m and put all the
Values in the equation on Y = c + m X and
Solve the forecast for the month of 2016 and 2017.
257 = 9c + 36m---(3) & 1232 = 36c + 204m ---(4)
Solving both the equation we get c = 14.96 & m = 3.4Y = 14.96 + 9 * 3.4 =45.56--forecast sales for
2016

20. Problem2. A survey revealed that the demand for coolers in towns has the
following data:
Fit a linear regression and estimate the demand for the cooler for a town whose
population is 20 × 106Population in towns in × 106;
X
5 7 8 11 14
n
0 1 2 3 4 (5)
No. of coolers demanded;
Y
45 65 55 75 95
As per the given problem I already defined the value of X,Y and n for clarity. Th
Solution as follows: Y = m X + C
X Y ΣXY ΣX2 ΣY= mΣX+ n C ΣXY= mΣX2 + C ΣX
5 45 225 25 Find the value
of
m and C from the above
two equations.
7 65 455 49
8 55 440 64 345=m*45+5*C 3275=m*455+ 45*C
11 75 825 121 m=3.4 C=38.4
14 95 1330 196 Y = m X + C
ΣX=45 ΣY=345 =3275 =455 Y=3.4*20+38.4=106.4
No. of cooler
required=106.4

21. Thank youTo
all