Interpolation and

UNIT 2 : INTERPOLATION AND Extrapolation using

Growth Rate Methods
EXTRAPOLATION USING
GROWTH RATE METHODS
Structure

2.1 Introduction

2.2 Why Interpolation and Extrapolation?

2.3 Distinguish Between Interpolation and Extrapolation

2.4 Assumptions

2.5 Interpolation and Extrapolation Explained With an Example

2.6 Methods of Interpolation and Extrapolation

2.6.1 Linear Interpolation/Extrapolation
2.6.2 Binomial Expansion Method
2.6.3 Arithmetic Interpolation and Extrapolation
2.6.4 Geometric Interpolation and Extrapolation

2.7 Application of Interpolation and Extrapolation

2.7.1 Estimating Intermediate Values
2.7.2 Missing Data
2.7.3 Non-availability of data
2.7.4 Bring Uniformity in the Data
2.7.5 Forecasting

2.8 Let Us Sum Up

2.9 References and Suggested Reading

2.10 Check Your Progress- Possible Answer

2.1 INTRODUCTION
Population study is dynamic in nature of research and population data is
changing over the period of time. Population data are providing by either
census operation or by periodical survey. Since most of the modern-day
measurements are discrete in nature, we normally do not have an analytic
function to demonstrate these set of measurement and interpolation and
extrapolation are useful to find that function. Interpolation is an important
statistical technique to estimate the unknown value from known given values
whereas extrapolation is the process of calculating unknown values beyond
the given data points. The central idea here is to find an analytic function that
passes through given points to interpolate or extrapolate for any arbitrary
point. For instance, we are given the steel production figures of a country for
the years 1931, 1941, 1951, 1961, 1971, 1981, 1991, 2011 etc., then to find
125
Estimation of out the production figure for any year falling between 1931 and 2011 can be
Population termed as interpolation. If we compute the production figure for any year
outside the interval 1931 to 2011 it will be termed as extrapolation. Keep in
mind that population is in a constant state of change, both in terms of size and
characteristics. We know that periodic census enumeration obtain data on the
size and composition of the population at the time of census. For many
practical purposes, it is important to know the number and characteristics of
people during the period between the two censuses and even after the last
census. That is why population interpolation and extrapolation are important
in population studies.

After reading this unit learner be able to -

I. Explain the basic concept of interpolation and extrapolation

II. Differentiate between interpolation and extrapolation

III. Apply different methods of interpolation and extrapolation

IV. Explore the application of interpolation and extrapolation

2.2 WHY INTERPOLATION AND

EXTRAPOLATION?
Sometimes, we are requiring mid-point of time data or data which is require
for future prediction. In this case we need both interpolated and extrapolated
data. Interpolation and extrapolation can be measured based on the growth
rate of the population.

In the mathematical, interpolation is a method of constructing new data

points within the range of a discrete set of known data points. In other words
interpolation is the approximation of a complicated function by a simple
function. Suppose we know the function but it is too complex to evaluate
efficiently. Then we could pick a few known data points from the
complicated function, creating a lookup table, and try to interpolate those
data points to construct a simpler function. Of course, when using the simple
function to calculate new data points we usually do not receive the same
result as when using the original function, but depending on the problem
domain and the interpolation method used the gain in simplicity might offset
the error.

Definition of Interpolation

Interpolation is a technique of obtaining the most likely estimate of certain

quality under certain assumptions (Ethanie). According to Hirsch
“interpolation is the estimations of most likely estimate in given conditions.
The technique of estimation a past figure is termed as interpolation; while
that of estimating a probable figure for the future is called extrapolation.”
W.M. Harper has defined it as “interpolation consists in reading a value
126
which lies between tow extreme points. Extrapolation means reading a value Interpolation and
Extrapolation using
that lies outside the two extreme points.” (https://www.google.com/search? Growth Rate Methods
client=firefox-b-d&q=interpolation+and+extrapolation+pdf).

Based on the definition we can say that interpolation and extrapolation are
the same techniques, but interpolation is related to the past trends whereas
extrapolation gives us the future trends of population figures or forecast or
number of events. Following objectives of interpolation and extrapolation-

- Estimating the size of population at the points of time between the

censuses or Intermediate Value

- Estimating the size of population at times during the first year or so

following the census.

- Adjusting the population counts or estimates to midyear or other

dates in the same or adjacent years.

Check Your Progress 1

Note: a) Write your answer in about 50 words

b) Check your answer with possible answers given at the end of the
unit.

1) Explain briefly Interpolation and Extrapolation ………………………

……………………………………………………………………...……
………………………………………………………………………...…
…………………………………………………………………………...

2.3 DISTINGUISH BETWEEN INTERPOLATION

AND EXTRAPOLATION
Both extrapolation and interpolation are useful methods to determine or
estimate the hypothetical values for an unknown variable based on the
observation of other data points. However, it can be hard to distinguish
between these methods and understand how they differ from each other
(https://www.techtarget.com/whatis/definition/extrapolation-and-interpolation).

Distinction between Interpolation and extrapolation

Interpolation Extrapolation
The reading of values between two points in a Estimating a value that's outside the data set
data set
Primarily used to identify missing past values Plays a major role in forecasting
The estimated record is more likely to be The estimated values are only probabilities,
correct. so they may not be entirely correct.
It is information for past data It is projected data for future
Source: https://www.techtarget.com/whatis/definition/extrapolation-and-interpolation

127
Estimation of One of the easiest ways to understand these differences is to understand the
Population prefix of each term. Extra- refers to "in addition to," while inter- means "in
between." Thus, extrapolation indicates a user is trying to find a value in
addition to existing values, while interpolation means that they want to
determine a new value in between existing value. The basic differences
between these two terms are given in above table.

2.4 ASSUMPTIONS
Despite the fact there are number of dependable methods of interpolation and
extrapolation to find out most possible values of the past and future but even
then, there are certain conditions that need to be met to get reliable outcomes.
Such as-

o Enough observations should be available based on which the

dependent variables are to be estimated.

o As much as possible the given values of x i.e., the independent

variable should be at equal interval. There should be no consecutive
missing term in the data set.

o The variables under consideration must be related to each other.

o The rate change of values in independent variable should be

uniform.

o The data under consideration should not be affected by external

factors.

We can say that interpolation and extrapolation are the population estimates
and population projection techniques. If the situation is somewhat different
for post-census estimates; that is in the case of estimates for dates subsequent
to the most recent census for which data is available. First, there is no later
census to indicate the upper limit of the range within which the true figures
usually lie. Second, post census estimates most commonly are made for dates
so recent that vital statistics and other data bearing on population change
during the period since the last census may not yet be available for the entire
period.

As we have read above that interpolation and extrapolation are estimates of

present and past and future figure of population or events. Perhaps the most
accepted method of estimating population which is strictly mathematical is
that which assumes a constant rate of population change or constant
increment to the population each year. Such estimates are more easily
prepared than those of any other type. However, the basic assumption almost
always proves to have been in error if any considerable period of time is
involved.

Thus, interpolation is the process of mathematically calculating the size of

128 population at a point of time in between two dates at which its size is
unknown. This may be done on the basis of either constant increments in Interpolation and
Extrapolation using
number (arithmetic interpolation) or constant rates of increase (geometric Growth Rate Methods
interpolation). Extrapolation is the same process for years beyond the two
known dates.

2.5 INTERPOLATION AND EXTRAPOLATION

EXPLAINED WITH AN EXAMPLE
Interpolation means determining a value from the existing values in each data
set. Another way of describing it is the act of inserting or interjecting an
intermediate value between two other values. In data statistics, interpolation
is about calculating a function's value based on the value of other data points
in a given sequence. This function may be represented as f(x), and the known
x values may range from X0 to Xn.

For example, suppose we have a regression line y = 30x + 40. We know that,
to produce this "best-fit" line, the value of x must between 0 and 10. Suppose
we choose x = 6. Based on this best-fit line and equation, we can estimate the
value of y as the following:
y = 30(6) + 40 = 220

Our x value (6) is within the range of acceptable x values used to make the
line of best fit, so this is a valid y value, which we have calculated by
interpolation.

Extrapolation on the contrary is about predicting hypothetical values that fall

outside a particular data set. The predictive quality of extrapolation means the
method is usually used to predict unknown future values, unlike
interpolation, which is usually about estimating past values.

For example, suppose a data set consists of four given values: 1, 3, 5 and 7. If
these values were plotted on a graph and the line was expected to continue in
the same way, the fifth value could be extrapolated as 9.

In population studies, interpolation and extrapolation can be measured by

growth rate of population between two censuses. Thus, exponential
population growth rate can be measure before interpolation. The following
formula have been used for the analysis of population growth and
interpolation and extrapolation of population in different years which in
unknown to us.
Population Growth Rate (1991 to 2001 Census) =0.1*LN (Population 2001/Population 1991)
Interpolation (1991 to 1996) = Population 1991*EXP (5.33*Growth rate 1991 to 2001)
Interpolation (2001 to 2005) = Population 2001*EXP (4.33*Growth rate 2001 to 2011)
Interpolation (2001 to 2010) = Population 2001*EXP (9.33*Growth rate 2001 to 2011)
Extrapolation (2011 to 2013) = Population 2011*EXP (12.33*Growth rate 2001 to 2011)
129
Estimation of We can observe in above formula that interpolation during 1991 to 1996 has
Population used 5.33, which indicates gap between 1991 to 1996 years. It means, census
was conducted in 1991 and we have interpolated population for 1996. We
have also found that in the formula in decimal figure 0.33 which means first
quarter of the year 1991, because census is releasing its data in the March of
the census year. Similar patterns have adopted in interpolated population in
2005 and 2010, and extrapolated population in 2013. In extrapolated
population 2013 has used 12.33, indicates gap between 2011 to 2013.
Therefore, very easy to calculate interpolated and extrapolated population
which is unknown to us for any required year or time. Table 2.1 clearly
shows, population, population growth rate and interpolated and extrapolated
population for given years by using above formula.

Table 2.1: Population (Mil), annual growth rate, interpolated and

extrapolated population for UK
Extrapolated
Population Interpolated Population
(Mill) Growth rate Population (Mill) (Mill)
1991 2011 1991-01 2001-11 1996 2005 2010 2012
Uttarakhand (UK) 7.11 10.09 0.018 0.017 7.82 9.15 9.97 12.26
Almora 0.61 0.62 0.003 -0.001 0.62 0.63 0.62 0.61
Bageshwar 0.19 0.26 0.027 0.004 0.22 0.25 0.26 0.27
Chamoli 0.33 0.39 0.013 0.006 0.35 0.38 0.39 0.42
Champawat 0.23 0.26 -0.002 0.015 0.23 0.24 0.26 0.31
Dehradun 1.03 1.70 0.022 0.028 1.16 1.45 1.67 2.33
Garhwal (Pauri) 0.67 0.69 0.004 -0.001 0.69 0.69 0.69 0.68
Hardwar 1.14 1.89 0.024 0.027 1.30 1.62 1.86 2.56
Nainital 0.57 0.95 0.028 0.022 0.67 0.84 0.94 1.23
Pithoragarh 0.42 0.48 0.010 0.004 0.44 0.47 0.48 0.51
Rudraprayag 0.20 0.24 0.013 0.006 0.21 0.23 0.24 0.26
Tehri Garhwal 0.52 0.62 0.015 0.002 0.56 0.61 0.62 0.64
Udham Singh Nagar 0.97 1.65 0.025 0.029 1.10 1.40 1.62 2.29
Uttarkashi 0.24 0.33 0.021 0.011 0.27 0.31 0.33 0.37
Source: Census of India

We can read different methods of interpolation and extrapolation of

population for any years below. There are many different interpolation
methods, some of which are described below. Some of the concerns to take
into account when choosing an appropriate algorithm are: How accurate is
the method? How expensive is it? How smooth is the interpolate? How many
data points are needed? The term extrapolation is used if we want to find data
points outside the range of known data points. As far as the methods are
concerned there is no difference between interpolation and extrapolation.

In curve fitting problems, the constraint that the interpolant has to go exactly
through the data points is relaxed. It is only required to approach the data
points as closely as possible (within some other constraints). This requires
parameterizing the potential interpolants and having some way of measuring
the error. In the simplest case this leads to least squares approximation.

Approximation theory studies how to find the best approximation to a given

function by another function from some predetermined class, and how good
130
this approximation is. This clearly yields a bound on how well the interpolant Interpolation and
Extrapolation using
can approximate the unknown function. Growth Rate Methods

Check Your Progress 2

Note: a) Write your answer in about 50 words

b) Check your answer with possible answers given at the end of the
unit.

1) Measure Population growth and Interpolation and Extrapolation ………

…………………………………………………………………………...
………...…………………………………………………………………
…………...………………………………………………………………
……………...……………………………………………………………

2.6 METHODS OF INTERPOLATION AND

EXTRAPOLATION
There have been numerous methods for interpolation and extrapolation in
statistics (Wahab, 2017), for instance, linear method, Newton’s advancing
difference method, Lagrange’s method, Binomial Expansion method, among
others. However, two most used and easily understandable interpolation/
extrapolation methods are the following:

2.6.1 Linear method

2.6.2 Binomial expansion method

Figure 1: Interpolation Figure 3: Extrapolation beyond the known values

2.6.1 Linear Interpolation/Extrapolation

Linear interpolation is among the simplest interpolation methods. Here, a
straight line is drawn between two points ( 0, 0) and ( 1, 1) on a graph to
determine the other unknown values(x, y). The simple method frequently
results in inaccurate estimates. Figure 1 show the interpolated value of x,
which is depicted in blue colour, within the range of known data points
shown in red colour. 131
Estimation of Similarly, figure 2 shows that extrapolation is used to predict the value of
Population function f(x) at an arbitrary data point x which is beyond the range input data
points and is demonstrated by a question mark.

2.6.2 Binomial Expansion Method

Also known as Taylor Polynomial, As can be seen from the definition of the
divided differences new data points can be added to the data set to create a
new interpolation polynomial without recalculating the old coefficients. And
when a data point changes we usually do not have to recalculate all
coefficients. Furthermore if the xi are distributed equidistantly the calculation
of the divided differences becomes significantly easier. Therefore the Newton
form of the interpolation polynomial is usually preferred over the Lagrange
form for practical purposes, although, in actual fact (and contrary to
widespread claims), Lagrange, too, allows calculation of the next higher
degree interpolation without re-doing previous calculations--and is
considerably easier to evaluate.

Built on the Binomial Theorem, this method is easy to understand and apply.
The basic requirement of this particular method is that the intervals of
dependent variables must be equal throughout the data and missing value, or
values of the independent variables must correspond to these intervals.

For instance, this method is applicable when the data looks like this-

Marks obtained 10 20 30 40 50
Number of students 5 23 ? 45 75

Following steps should be followed for estimating missing values by this

method:

o Find the number of known values (n) of the dependent variable Y.

o Then find ∆ =0 or 1 = 0.

o Expand 1 = 0 according to binomial expansion method.

Substitute the values and obtain one missing value.

o To get another missing value, use the same equation by doing small
alteration that is raising y suffix by 1.

o Substitute the values and obtain the second missing value.

Problem: Estimate the steel production for the year 2001 and 2005 from the
following data.

Year 1995 1997 1999 2001 2003 2005

Steel production (MT) 20 40 70 ? 130 ?

132
Solution: Let X be the year and Y be the steel production. Interpolation and
Extrapolation using
Growth Rate Methods
X 1995 1997 1999 2001 2003 2005
Y 20 (y0) 40 (y1) 70 (y2) ? (y3) 130 (y4) ? (y5)
The number of known values is n= 4.

∆ =0 or 1 =0

1 = 0.

1 =y4 – 4y3 + 6y2 – 4y1 + y0 = 0.

=130 –4y3 + 6(70) – 4(40) + 20 = 0

=130 – 4y3 + 420 – 160 + 20 = 0

=410 – 4y3 = 0

= -4y3 = -410

=y3 = 102.5

To find another value use

1 = y4 – 4y3 + 6y2 – 4y1 + y0 = 0.

Raise the suffixes by 1

y5 – 4y4 + 6y3 – 4y2 +y1 = 0

y5 – 4(130) + 6(102.5) – 4(70) + 40 = 0

y5 – 520 + 615 – 280 +40 = 0

y5 – 145 = 0

y5 = 145.

2.6.3 Arithmetic Interpolation and Extrapolation

In this interpolation, the difference between the sizes of population at two

different dates is prorated on the basis of the time interval between the earlier
known date and the estimate date. Thus, to arrive at the population of India in
2010, the different of population figure in census of India in 2001 to 2011
and estimate figure of population for 2010. You can see example in Table
2.1.

Similarly, arithmetic extrapolation is accomplished in the following three

steps-

133
Estimation of I- Finding the average increment per year by dividing the total
Population increase in population by the number of year between the last two
censuses.

II- Find the estimated increment upto the date of estimate by

multiplying the average annual increment by the number of years
from the last census date to the date of estimate.

III- Finding the extrapolated population on the estimate date by adding

that increment to the population at the last census date.

2.6.4 Geometric Interpolation and Extrapolation

In geometric interpolation the average annual rate of increase between two
dates for which the population is known is determined, and that rate is then
applied to estimate the population at some point of time between the two
dates. The calculation required to estimate the population of India in 2010 on
the basis of the average annual rate of increase between 2001 and 2011
censuses are as follows-

Average annual rate of increase (Population growth rate), 2001 to 2010:

log p (2011)-log p(2001)

log(1+r) =
n(10)

Population in 2010 if the population growth rate had been constant from 2001
to 2011:

log P (2010)= log p(2001) + n log (1+r)

the same procedures are used for extrapolation as for interpolation. The only
difference between the two, as explained earlier, is that interpolation in
between two known values and extrapolation is for years later than the period
between the two known values.

2.7 APPLICATION OF INTERPOLATION AND

EXTRAPOLATION
Interpolation and extrapolation techniques are the foundation stone of many
widely used applications in the fields of mathematics, statistics, science,
economics, and engineering. Such as:

2.7.1 Estimating Intermediate Values

Interpolation is useful in determining the intermediate values from the given
data, which if otherwise collected may need much of efforts. Three are a
number of figures or data which are collected at some fixed intervals. There
is no provision to collect the same data for any other period. In that case only
interpolation tool come to our rescue. For instance, Indian Census collects
134
population data every ten-year interval. In case one needs population figure Interpolation and
Extrapolation using
for any inter-censual period, interpolation has to be used for the same. Simply Growth Rate Methods
put, we can say that the interpolation techniques help in filling the data gaps.

2.7.2 Missing Data

Many a time it happens that the collected data is misplaced, destroyed or lost
for some reasons like lack of seriousness, fire, war or flood etc. it is not
always possible to collect data again. The missing figures can be estimated
with the help of suitable interpolation methods.

2.7.3 Non-availability of data

Often it becomes practically impossible to collect the data at some particular
time or at some particular place. Then the data for the particular place or time
can be estimated on the basis of the other part of the given data. For instance,
it may become impossible to reach some geographic regions to collect
population data during census due to certain adversity. In such a situation,
one can estimate the population figure of that particular area on the basis of
the population figure for other years.

2.7.4 Bring Uniformity in the Data

The primary objective of data collection is to make comparisons. Suppose we
are given the number of students of a class ‘A’ who have got marks less than
5, 15, 25, 35………. etc. and the students of a class ‘B’ who have obtained
marks less than 10, 20, 30 ……etc. now to compare them properly both the
data must be in similar intervals. To solve these issues one can, utilize the
interpolation.

2.7.5 Forecasting
Extrapolation is the best possible tool for forecasting. All economic planning,
policy formulations etc. are nothing but future estimations. The methods
provide scientifically based formulae for predictions and forecasting.

Check Your Progress 3

Note: a) Write your answer in about 50 words

b) Check your answer with possible answers given at the end of the
unit.

1) Describe Application of Interpolation and Extrapolation ………………

……………………………………………………………………...……
………………………………………………………………………...…
…………………………………………………………………………...

135
Estimation of
Population
2.8 LET US SUM UP
In this unit we have learnt that interpolation and extrapolation are very
powerful statistical tools that are useful for estimating missing number in a
given data series and forecasting a number outside a given data points. We
also discussed the basic concepts, assumptions and methods involved in
interpolation and extrapolation, followed by their real-world application in
various fields. We have therefore dealt with the subject adequately. Solving
an interpolation problem leads to a problem in linear algebra where we have
to solve a system of linear equations. Using a standard monomial basis for
our interpolation polynomial we get the very complicated Vander monde
matrix. By choosing another basis, the Newton basis, we get a system of
linear equations with a much simpler lower triangular matrix which can be
solved faster.

2.9 REFERENCES AND SUGGESTED READING

Das, N. G. (1991). Statistical Methods in Commerce. Accountancy and
Economics, m. Das & co., Kolkata. Part–pp, 1-25.
Interpolation, Approximation And Extrapolation Course Notes Dept Of
Biomedical Engineering Kings College London,United Kingdom
Mandal, R.B., Uyanga, J and Prasad, H (2007), Introductory Methods in
Population Analysis, Concept Publishing Company, Mohan Garden, New
Delhi, pp, 119-120.
Steven C. Chapra (2007), “Applied Numerical Methods with MATLAB® for
Engineers and Scientists”, 3rd Edition, New York: McGraw-Hill, 17-18
Techmint: (Retrieve from https://www.techtarget.com/whatis/definition/
extrapolation-and-interpolation#:~:text=Extrapolation%20refers%20to%20
estimating%20an,within%20a%20sequence%20of%20values, accessed on
2nd March, 2023).
Wahab, M. A. (2017). Interpolation and Extrapolation. Proc. Topics Syst.
Eng. Winter Term,17,1-6.
https://www.google.com/search?client=firefox-b-d&q=interpolation+and+
extrapolation+pdf
William H. Press, Saul A. Teukolsky, William T. Vetterling, and Brian P.
Flannery (2007), “Numerical Recipes: The Art of Scientific Computing”, 3rd
Edition, Cambridge University Press
https://en.wikipedia.org/wiki/Bilinear_interpolation
http://www.cambridgeincolour.com/tutorials/image-interpolation.htm
https://en.wikipedia.org/wiki/Linear_interpolation
https://en.wikipedia.org/wiki/Extrapolation
https://de.mathworks.com/help/matlab/ref/interp1.html
136
 Interpolation and
2.10 CHECK YOUR PROGRESS- POSSIBLE Extrapolation using
ANSWER Growth Rate Methods

Check Your Progress 1

Q.1 Explain briefly Interpolation and Extrapolation

A- Interpolation is the process of mathematically calculating the size of

population at a point of time in between two dates at which its size is
unknown. This may be done on the basis of either constant increments in
number (arithmetic interpolation) or constant rates of increase
(geometric interpolation). Extrapolation is the same process for years
beyond the two known dates.

Check Your Progress 2

Q.2 Measure Population growth and Interpolation and Extrapolation

A- Following measure can be used to solve the problems

Population Growth Rate (1991 to 2001 Census) =0.1*LN (Population 2001/Population 1991)

Interpolation (1991 to 1996) = Population 1991*EXP (5.33*Growth rate 1991 to 2001)

Interpolation (2001 to 2005) = Population 2001*EXP (4.33*Growth rate 2001 to 2011)

Interpolation (2001 to 2010) = Population 2001*EXP (9.33*Growth rate 2001 to 2011)

Extrapolation (2011 to 2013) = Population 2011*EXP (12.33*Growth rate 2001 to 2011)

Check Your Progress 3

Q.3 Describe Application of Interpolation and Extrapolation

A- Interpolation and extrapolation techniques are the foundation stone of

many widely used applications in the fields of mathematics, statistics,
science, economics, and engineering. Application of Interpolation and
Extrapolation can be seen in 2.7 above.

137