MODULE On Index Numbers – 2025-26 (I)

Mathematical and
Statistical Methods
ECON F213

Dr. Rahul Arora (IC)

Assistant Professor,
Department of Economics & Finance,
BITS Pilani, Pilani Campus
[Link]@[Link]
Mob: +91 – 7607481292

Background design is taken from the presentation slides of Salvatore:

International Economics, 10th Edition © 2013 John Wiley & Sons, Inc.
 Introduction
❑ “Index Numbers are devices for measuring differences in the
magnitude of a group of related variables“ Croxon & Cowden

❑ Statistical measure designed to show changes in a variable or

group of variables over time or location or any other
characteristics

❑ An index computed from a single variable is called univariate

index

❑ An index constructed from a group of variables are known as

composite index

2
Classification of Index Numbers

❑ Price Index – compares the level of prices over time

❑ Quantity Index – measures the changes in quantity

over time

❑ Value Index – measure the changes in the rupee value

of a variable

❑ Special purpose index – measure the changes in

particular characteristics of a variable designed for a
special inquiry

3
Steps in Constructing Index
Numbers

1. Deciding purpose of the making an index

2. Selection of a base period
3. Selection of number of items – sample size, variables
4. Price quotations
5. Choice of an average
6. Selection of appropriate weights
7. Selection of an appropriate formula

4
 Methods of Constructing Index
Numbers

1. Un-Weighted indices

❑ Simple aggregative

❑ Simple average of relatives

2. Weighted indices

❑ Weighted aggregative

❑ Weighted average of relatives

5
 Unweighted Index Numbers

Simple Aggregative Method –

❑ Simplest method of constructing index numbers

❑ Required data – Prices of commodities at two points in

time

σ 𝑃1
𝑃01 = × 100
σ 𝑃0

Where P1 is the current year prices of various

commodities and P0 is the base year prices for the same
commodities 6
 Unweighted Index Numbers

Simple Average of Price Relatives Method –

❑ Simple method of constructing index numbers

❑ Required data – Prices of commodities at two points in

time

𝑃1
σ × 100
𝑃0
𝑃01 =
𝑁

Where N is the number of items whose price relatives are

averaged
7
 Weighted Index Numbers

Weighted Aggregative Index –

❑ Same as simple aggregative method with a difference

of weights

❑ Few important methods are –

1. Laspeyres method overestimates(leaves an upward bias)

2. Paasche method underestimates (downward bias)
3. Dorbish and Bowley’s method
4. Fisher’s Ideal method
5. Marshall-Edgeworth method
6. Kelly’s method
8
 Laspeyres Method

❑ Weights are assigned by quantities to the base period

❑ It answers: “What is the change in the aggregate value

of the base period list of goods when valued at given
period prices”

σ 𝑝1 𝑞0
𝑃01 = × 100
σ 𝑝0 𝑞0

Disadvantage – doesn’t take into consideration the

consumption pattern; upward bias

9
 Paasche’s Method

❑ Weights are assigned by quantities in the current year

❑ It answers: “What would be the value of the given

period list of goods when valued at base year prices”

σ 𝑝1 𝑞1
𝑃01 = × 100
σ 𝑝0 𝑞1

Disadvantage – costly to collect data on new goods for the

base year; downward bias

10
 Dorbish and Bowley’s Method

❑ Arithmetic mean of the two indices (L & P) computed

above

❑ It takes into account the influence of both the periods

𝐿+𝑃
𝑃01 = × 100
2

Where –
L is Laspeyres Index and P is Paasche index

11
 Fisher’s Ideal Index

❑ This is the ideal formula

❑ Geometric mean of the L & P indices

𝑃01 = 𝐿 × 𝑃

Where –
L is Laspeyres Index and P is Paasche index

12
 Fisher’s Ideal Index – Why Ideal ?

❑ Based on the Geometric Mean

❑ Takes into account current as well as base year prices

❑ Satisfies both time reversal and factor reversal tests

❑ Free from all biases (upward and downward biases

would be cancel out)

13
 Marshall-Edgeworth Method

❑ Considers both current as well as base year prices and

quantities

❑ Give results very close to ideal measure

σ 𝑝1 (𝑞0 + 𝑞1 )
𝑃01 = × 100
σ 𝑝0 (𝑞0 + 𝑞1 )

14
 Kelley’s Method

❑ Weights are quantities (can be of any year)

❑ Quantity can be average quantity of two or more years

❑ Also known as fixed weight aggregative index

σ 𝑝1 𝑞
𝑃01 = × 100
σ 𝑝0 𝑞

In case of two year’s average then quantity can be

calculated as –
𝑞0 + 𝑞1
𝑞=
2 15
 Weighted Index Numbers

Weighted Average of Relatives –

❑ Important methods are –

1. Weighted arithmetic mean of relatives index

2. Weighted geometric mean of relatives index

16
 Weighted arithmetic mean of
relatives index
Steps to Construct –
❑ Obtain percentage relatives for each period

❑ Multiply with the value weights

❑ Calculate the weighted arithmetic mean of weighted

relatives as follows –
σ 𝑃𝑉
𝑃01 =
σ𝑉

Where –
P is the price relatives
V is value weights 17
 Weighted geometric mean of
relatives index
Steps to Construct –
❑ Obtain percentage relatives for each period and find
logarithm of each

❑ Multiply logarithm with the value weights

❑ Divide the total by weights and take antilog of quotient

obtained from follows –

σ 𝑉 (𝐿𝑜𝑔 𝑃)
𝑃01 = 𝐴𝑛𝑡𝑖𝑙𝑜𝑔
σ𝑉
Where –
P is the price relatives
V is value weights
18
 Quantity Index Numbers

❑ Used to indicate the real level of output in the country

❑ Take price or value as weights

❑ Replace quantity with price in the price indices

formulae (Laspeyre, Paasche, and Fisher)

19
 Value Index Numbers

❑ No need of taking weights

❑ Weights are inherent

Formula of calculating –
σ 𝑝1 𝑞1
𝑉= 𝑋 100
σ 𝑝0 𝑞0

σ 𝑉1
𝑉= 𝑋 100
σ 𝑉0

Where V is the value (Price X Quantity)

20
 Adequacy Tests

❑ Unit Test
❑ Time Reversal Test
❑ Factor Reversal Test
❑ Circular Test
For detailed contents, refer theoretical notes

21
 Base Shifting
❑ If base becomes old, shift the base period to get more
accurate results

❑ One possibility is to re-compute all index numbers using

new base period

❑ Another possibility is directly shift the base using base

shifting method

❑ Divide all the index numbers for various years

corresponding to old base period with the index numbers
corresponding to new period and expressing the results as
percentages
Note – It may not coincide with the value obtained by calculating entirely the new
index on a same new base year because of different weighing system
22
 Splicing
❑ When Index number series is available for a period of time,
then undergoes substantial revision including a shift in the
reference period then one can use Splicing

Note – It may not coincide with the value obtained by calculating entirely the new
index on a same new base year because of different weighing system

23
 Index Numbers in Deflating

❑ Purchasing power of the money can be calculated

through Index Numbers

❑ Purchasing power is the reciprocal of the price index

❑ Real wage can be calculated by multiplying money

wage with the purchasing power of that money

❑ Numerical

24
 Chain Index Numbers

❑ Numerical

❑ Fixed and Chain Index – Similarities and Differences

25
 Applications of Index Numbers

❑ Consumer Price Indices (Cost of living index)

❑ GDP Deflator

❑ Wholesale Price Index

❑ Index of Industrial Production (IIP)

❑ Calculation of Inflation

26
Reference

(Statistics) – Gupta, S.P., Statistical Methods, Sultan

Chand and Sons, 45th Revised Edition (2017)

27