STATISTICS | 2K17/ECO/48

STATISTICS
Assignment title: INDEX NUMBER

Submitted to: SIR SIKANDAR SOOMRO

 STATISTICS | 2K17/ECO/48

INDEX NUMBER
INTRODUCTION
An index number is a statistical derives to measure
changes in the value of money. It is a number which represents the
average price of a group of commodities at a particular time in relation
to the average price of the same group of commodities at another time.

Professor Chandler defines it thus: “An index number of prices is a

figure showing the height of average prices at one time relative to their
height at some other time that is taken at the base period.” To
understand the meaning of the term index number, three points are to
be noted.

First, an average figure relates to a single group of commodities. But

the various items in the group are expressed in different units. For
example, a consumer price index contains such diverse items as food,
clothing, fuel and lighting, house rent, and miscellaneous things. Food
consists of wheat, ghee, etc. expressed in kg cloth is expressed in
meters, and lighting in KWS.

An index number expresses the average of all such diverse items in

different units. Second, an index number measures the net increase or
decrease of the average prices for the group under study. For instance,
if the consumer price index has increased from 150 in 1982 as
compared to 100 in 1980, it shows a net increase of 50 per cent in the
prices of commodities included in the index. Third, an index number
measures the extent of changes in the value of money (or price level)
1|Page
 STATISTICS | 2K17/ECO/48

over a period of time, given a base period. If the base period is the year
1970, we can measure change in the average price level for the
preceding and succeeding years.

Methods of Construction of Index Number:

In constructing an index number, the following steps should be noted:

1. Purpose of the Index Number

Before constructing an index number, it should be decided the
purpose for which it is needed. An index number constructed for one
category or purpose cannot be used for others. A cost of living index of
working classes cannot be used for farmers because the items entering
into their consumption will be different.

2. Selection of Commodities:
Commodities to be selected depend upon the purpose or objective of
the index number to be constructed. But the number of commodities
should neither be too large nor too small.

Moreover, commodities to be selected must be broadly representative

of the group of commodities. They should also be comparable in the
sense that standard or graded items should be taken.

3. Selection of Prices:
The next step is to select the prices of these commodities. For this
purpose, care should be taken to select prices from representative
persons, places or journals or other sources. But they must be reliable.
Prices may be quoted in money terms i.e. Rs. 100 per quintal or in
quantity terms, i.e. 2 kg per rupee. Care should be taken not to mix

2|Page
 STATISTICS | 2K17/ECO/48

these prices. Then the problem is to select wholesale or retail prices.

This depends on the type of index number. For a consumer price
index, wholesale prices are required, while for a cost of living index,
retail prices are needed. But different prices should not be mixed up.
4. Selection of an Average:
Since index numbers are averages, the problem is how to select an
appropriate average. The two important averages are the arithmetic
mean and geometric mean. The arithmetic mean is the simpler of the
two. But geometric mean is more accurate. However, the average
prices should be reduced to price relatives (percentages) either on the
basis of the fixed base method or the chain base method.

5. Selection of Weights:
While constructing an index number due weightage or importance
should be given to the various commodities. Commodities which are
more important in the consumption of consumers should be given
higher weightage than other commodities. The weights are determined
with reference to the relative amounts of income spent on
commodities by consumers. Weights may be given in terms of value or
quantity.

6. Selection of the Base Period:

The selection of the base period is the most important step in the
construction of an index number. It is a period against which
comparisons are made. The base period should be normal and free
from any unusual events such as war, famine, earthquake, drought,
boom, etc. It should not be either very recent or remote.

3|Page
 STATISTICS | 2K17/ECO/48

7. Selection of Formula:
A number of formulas have been devised to construct an index
number. But the selection of an appropriate formula depends upon the
availability of data and purpose of the index number. No single
formula may be used for all types of index numbers.

EXAMPLE

We give below an example each of the simple price index and the
weighted price index.

Simple Price Index:

To construct a simple price index, compute the price relatives and
average them. Add the price relatives and divide them by the number
of items. Table 64.1 illustrates the construction of a simple index of
wholesale prices.

Commodity Prices in Base Prices in Price

1970(P0) 1970=100 1980(P1) = Relatives
P1/P0xl00 (R)
A Rs . 20 per kg 100 Rs. 25 125
В 5 per kg 100 10 200
С 15 per metre 100 30 200
D 25 per kg 100 30 120
E 200 per quantal 100 450 225
N=5 500 ∑R = 870
TABLE 64.1
Price index in 1980 = Prices in 1980 / Prices in 1970 x 100

Or ∑P1/P0 x 100 = 870/500 x 100 = 174

4|Page
 STATISTICS | 2K17/ECO/48

Using arithmetic mean, price index in 1980 = ∑R/N =

870/5 = 174

The preceding table shows that 1970 is the base period and 1980 is the
year for which the price index has been constructed on the basis of
price relatives. The index of wholesale prices in 1980 comes to 174.
This means that the price level rose by 74 per cent in 1980 over 1970.

Weighted Price Index:

Taking the example of Table 64.2 already given, we assign high
weights to commodities of greater importance to consumers.

Commodity Weight Prices Base Prices Price WxR

(W) in 1970 in Relatives
1970 = 1980
Rs 100 Rs (R)

A 6 20 100 25 125 750

В 4 5 100 10 200 800

С 2 15 100 30 200 400

D 4 25 100 30 120 480

E 10 200 100 450 225 2250

∑24 ∑WR =
4680

5|Page
 STATISTICS | 2K17/ECO/48

TABLE 64.2
Using arithmetic mean, the weighted price index in 1980 = 4680/24 =
195.

The weighted price index is more accurate than the simple price index.
In the example given above, the weighted price index shows an
increase of 91 per cent in the price level in 1980 over 1970 as against
the increase of 74 per cent according to the simple price index.

THE END

6|Page
 STATISTICS | 2K17/ECO/48

7|Page