Time series Analysis and Forecasting

Time series:
A time series is a set of observation taken at a specific time, usually at equal intervals.

Mathematically, a time series is defined by the values

Y 1 ,Y 2 ,... of a variable Y (temperature, closing price of

share, etc.) at times

t1 ,t 2 ,... thus Y is a function of time (t); this is symbolized by Y =f ( t ) .

Components of time series or Variations in time series:

The characteristics movement of time series may be classified into four main types, called the components of a
time series. There are four kinds of change or variation involved in time – series analysis:
1. Secular trend
2. Cyclical fluctuation
3. Seasonal variation
4. Irregular variation.

The first type of change secular trend, the value of the variable tends to increase or decrease over a long period of
time.
Figure 1(a) shows a secular trend in an increasing but fluctuating time series.

The second type of variation seen is a time series is cyclical fluctuation. The most common example of cyclical
fluctuation is the business cycle.
Figure 1(b) illustrates a typical pattern of cyclical fluctuation above and below a secular trend line. Also note that
cyclical fluctuations do not follow any regular pattern but move in a somewhat unpredictable manner.

The third kind of change in time series “seasonal variation” involves patterns of change with in a year that tends to
be repeated from year to year.
Figure 1(c) Displays seasonal variation. Notice how it peaks in the fourth quarter of each year.
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

Figure 1: Time series

variations

Irregular variation is the forth type of change in time series caused by entirely unpredictable reasons and
occurred in totally unpredictable manner.
Figure 1(d) Illustrates irregular variation.

Translating or Coding of time - One general steps involved in time series:

In most of the cases translating or recoding of the given time values is required for computational convenience.
The necessary steps of translating time are discussed below:

1. When there are odd numbers of time point in 2. When there are even numbers of time point in
the data series the procedure will be as follows: the data series the procedure will be as follows:

Page 2 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

First determine X̄ =1992 (In this case) First determine X̄ =1992.5 (In this case)
Year Translated or Coded time Translated or
X̄ Year
Coded time
(X) x=( X − X̄ ) ( X − X̄ )
(X)
1989 1989 – 1992 = -3 x=( X − X̄ )∗2
1990 1990 – 1992 = -2
1989 1989 - 1992.5 = -3.5 -7
1991 1991 – 1992 = -1
1990 1990 - 1992.5 = -2.5 -5
1992 1992 1992 – 1992 = 0
1991 1991 - 1992.5 = -1.5 -3
1993 1993 – 1992 = 1
1992 1992 - 1992.5 = - 0.5 -1
1994 1994 – 1992 = 2
1993 1993 - 1992.5 = 0.5 1
1995 1995 – 1992 = 3
1994 1994 - 1992.5 = 1.5 3
1995 1995 - 1992.5 = 2.5 5
1996 1996 – 1992.5 = 3.5 7

Trend Analysis:
Of the four components of a time series, secular trend represents the long-term direction of the series. One way to
describe the trend component is to fit a line visually to a set of points on a graph. Any given graph, however, is
subject to slightly different interpretations by different individuals. We can also fit a trend line by the method of
least square.

Trend line takes different form. It can be linear or curve linear

Reasons for studying Trends:
1. The study of secular trends allows us to describe historical pattern.
2. Studying secular trends permits us to project past pattern, or trends, in to the future.
3. In many situations, studying the secular trend of a time series allows us to eliminate the trend component
from the series.

Fitting the linear trend by the Least – Square Method:

Problem 1: Otobi is a furniture manufacturing company. Since they started the company, the number of tables
they have sold is represented by this time series:

Year 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Table
42 50 61 75 92 111 120 127 140 138
Sold:

1. Find the linear equation that describes the trend in the number of tables sold by Otobi.
2. Estimates their sales of tables in 1998.

Solution:
Follow the Steps below for solving the problem:

Page 3 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

Step 1: Find the recoded time as mentioned before and then complete / construct the table below:
Year
2
x Y xY x 956
(X) a=Ȳ = =95. 6
1987 -9 42 - 378 81 1. 10
1988 -7 50 - 350 49
1989 -5 61 25 b=
∑ xY = 1978 =
1990
1991
-3
-1
75
92
∑ x 2 330
1992 1 1 Y^ =95. 6+5. 9939 x
1993 3 9
1994 5
2. Y^ =95.6+5.9939(13)=173.5
1995 7 140 980
1996 9 138 1242 81 Because for 1998: x= 7.5*2=13
0 956 1978 330

Problem 2: The number of faculty owned computers at the university of Ohio increased dramatically between
1990 and 1995:

Year 1990 1991 1992 1993 1994 1995

Number
50 110 350 1020 1950 3710
of PC

a. Develop a linear estimating equation that best describes the data.

b. Develop a 2nd – degree-estimating equation that best describe the data.
c. Estimate the number of PCs that will be in use at the university in 1999, using both equations.

Solution:
Follow the Steps below for solving the problem:

Step 1: Find the recoded time as mentioned before and then complete / construct the table

Year
2 2 4
x Y xY x xY x
(X)
1990 -5 50 - 250 25 1250 625
1991 -3 110 - 330 9 990 81
1992 -1 350
1993 1 1020
1994 3 1950 17550 81
1995 5 3710 18550 25 92750 625
0 7190 24490 113910 1414

Calculation for

a. a=Ȳ =

Page 4 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

b=
∑ xY =
∑ x2
Y^ =a+bx=
b. Equations to determine the 2nd degree estimating equation that best describe the data are

∑ Y =na+c ∑ x 2 ⇒ 7190=6 a+70 c

∑ x 2 Y = a ∑ x 2+c ∑ x 4 ⇒113910=70 a+1414 c
Solving these equations we get:

a=611.8750 c=502679
2
Y^ =611. 8750+349 . 8571 x +50 .2679 x

c.

Linear forecast: ⇒ Y^ =1198.33+349.8571(13) =

nd
2 degree equation forecast: ⇒ Y^ =611. 8750+349. 8571(13 )+50 . 2679(13)2 =
Problem 3: (Assignment)
Jeff Richardson invested his life savings and began a part time carpet cleaning business in 1986. Since 1986, Jeff’s
reputation has spread and business has increased. The average numbers of homes has cleaned per month each
year are:

Year 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Turn
6.4 11.3 14.7 18.4 19.6 25.7 32.5 48.7 55.4 75.7 94.3
over

a. Develop a linear estimating equation that best describes the data.

b. Develop a 2nd – degree-estimating equation that best describe the data.
c. Estimate the turn over 1999, using both equations.
Problem 4:
Listed below are the net sales (in million taka) for RAK ceramic industries from its inception in 2000 to 2011.
Year Net sales Year Net sales
2000 9.23 2006 38.43
2001 12.47 2007 45.73
2002 15.47 2008 53.55
2003 19.53 2009 58.24
2004 24.15 2010 64.81
2005 30.21 2011 73.09
For the given data
i. Determine the least squares equation.
ii. Determine the estimated sales for 2012 and 2013.
Problem 5:

Page 5 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

a. Listed below are the net sales for the pharmaceutical company for the six years from 2002 to 2009. The net
sales are in millions of taka.
Year Net Sales (in million Tk.) Year Net Sales (in million Tk.)
2004 6714 2008 9762
2005 7991 2009 10180
2006 9075 2010 8334
2007 9775 2011 8272
For the given data
iii. Determine the least squares equation.
iv. Determine the estimated sales for 2012 and 2013.
v. Estimate the trend values.
vi. Graphically show the original trend.

Cyclical Variation:
Cyclical variation is the component of a time series that tends to oscillate above and below the secular trend line
for periods longer than 1 year. The procedure used to identify cyclical variation is the residual method.
Residual Method:

The usual process or steps to find the cyclical variation using the residual method is discussed here along with an
example:
Problem 1:
The western Natural Gas Company has supplied 18, 20, 21, 25 and 26 billion cubic feet of gas, respectively for the
years 1991 to 1995.
a. Find the linear estimating equations that best describe these data.
b. Calculate the percent of trend for these data.
c. Calculate the relative cyclical residual for these data.
d. In which years does the largest fluctuation from trend occur, and is it same for both methods?

Solution: Follow the steps below for required solution:

Step 1: Complete the following table:

Year Y
2
x Y xY x Y^ ×100 Y −Y^
(X) Y^ ×100
Y^
1991 -2 18 - 36 4 17.8 101.12 1.12
1992 -1 20 - 20 1 19.9 100.50 0.50
1993 0 21
1994 1 25
1995 2 26 52 26.2 99.24 - 0.76
0 110 21

Step 2: Calculate the following:

Page 6 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

a. a=Ȳ =

b=
∑ xY =
∑ x2
Y^ =
b. See the last column above for percentage of trend.
c. See the last column above for relative cyclical residual.
d. Largest fluctuation (by both methods) was in 1993.

Seasonal Variation:

Besides secular trend and cyclical variation, a time series also include seasonal variation. Seasonal variation is
defined as repetitive and predictable movement around the trend line in one year or less. In order to detect
seasonal variation, time intervals must be measured in small units, such as days, weeks, months or quarters.

Three reasons for studying seasonal variation are:

1. We can establish the pattern of past changes.

2. It is useful to project past patterns in to the future.
3. Once we have established the seasonal pattern that exists, we can eliminate its effects from the time
series.

Ratio to Moving Average Method is popularly used to determine the seasonal variation.

Irregular Variation is the final component of a time series. After we have eliminated trend, Cyclical and Seasonal
variations from a time series we still have an unpredictable factor left. Typically, irregular variation occurs over
short intervals and follows a random pattern. Because of the unpredictability of irregular variation we do not
attempt to explain it mathematically.

Page 7 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

Index Number

Index number is defined as an indicator of average percentage change in a series of figures where one figure
(called the base) is assigned an arbitrary value of 100, and other figures are adjusted in proportion to the base1.

An index number is a device which shows by its variation he changes in a magnitude which is not capable of
accurate measurement in it or of direct valuation in practice 2. It represents the general level of magnitude of the
changes between two or more situations of a number of variables taken as a whole 3. Index numbers are
quantitative measures of the general level of growth of prices, production, inventory and other quantities of
economic interest.

An index number may be described as a specialized average designed to measure change in the level of
phenomenon with respect to time, geographic location or other characteristics such as income etc. Thus when we
say that the index number of wholesale prices is 125 for the period January 2010 compared to January 2009, it
means there is a net increase of whole commodities to the extent of 25 percent.

The following points are worth considering for a proper understanding of the term index number:
i. Index numbers are specialized averages.
ii. Index numbers measures the change in the level of a phenomenon
iii. The index number measures the effect of changes over a period of time.

Uses of Index Number:

i. They help in framing suitable policies.
ii. They reveal trends and tendencies.
iii. Index numbers are very useful if deflating.

Classification of Index Numbers

Index number may be classified in terms of what measure. In economics and business the classifications are:
i. Price index number ii. Value index number
iii. Quantity index number iv. Special purpose index number.

Methods of Constructing Index Numbers:

A large number of formulae have been devised for constructing index numbers. Broadly speaking they can be
grouped under two heads:
i. Un-weighted indices.
ii. Weighted indices

1
http://www.businessdictionary.com/definition/index-number.html. Accessed on May 31, 2010.
^ .019
¿2
anti log 10 ¿
2
Wheldom: Business Statistics.
^
anti log 10¿2 .019 ¿
3
Karmel: cited in S. P. Gupta and M.P. Gupta.
^ .019
¿2
anti log 10 ¿

Page 8 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

In the un- weighted indices, weights are not expressly assigned whereas in the weighted indices, weighted are
assigned to the various items. Each of these types may further be divided under two heads:
a. Simple aggregative b. Simple average of price relatives

The following charts illustrates the various methods

Index
Number

Un-
Weighted
weighted

Simple Simple
Simple average of Simple average of
Aggregative price Aggregative price
relatives relatives

i. Un-weighted Index Numbers

a. Simple Aggregative Method
This is the simplest method of constructing index numbers. When this method is used to construct a price
index, the total of current year prices for the various commodities in question is divided by the total of base year
price and the quotient is multiplied by 100. Symbolically,

P01=
∑ P1∗100
∑ P0
Where

∑ P1 = Total of current year prices for various commodities, and

∑ P0 = Total of base year prices for various commodities

This method of constructing the index is very simple and the steps required in computation are:

a) Add the current year prices for various commodities, i.e., obtain ∑ P1 .
b) Add the base year prices for the same commodities, i.e., obtain ∑ P0 .
c) Divide ∑ P1 by ∑ P0 and multiply the quotient by 100.

Illustration 1:
From the following data construct an index number for 2008 taking 2007 as base:

Commodity and Unit Price (tk.) in 2007 Price (tk.) in 2008

Butter (kg.) 110.000 120.00
Cheese (kg.) 75.00 80.00
Milk (lt.) 13.00 13.00
Bread (lb.) 9.00 9.00
Eggs (Doz.) 18.00 20.00
Ghee (1 Can) 850.00 860.00

Page 9 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

Solution:

Commodity and Price (tk.) in 2007 Price (tk.) in 2008

Unit ( Po ) ( P1 )
Butter (kg.) 110 120
Cheese (kg.) 75 80
Milk (lt.) 13 13
Bread (lb.) 9 9
Eggs (Doz.) 18 20
Ghee (1 Can) 850 860
∑ P0 =1075.00 ∑ P1=1102.00
P 01=
∑ P1 ∗100= 1102 ∗100=102.51
∑ P0 1075
This means that as compared to 2007, in 2008 there is a net increase in price of commodities included in the index
to the extent of 2.51%.

Limitations of the method:

There are two main limitations of the simple aggregative index:
a) The units in which prices of commodities are given affect the price index.
b) No importance is given to the relative importance of the commodities.

b. Simple Average of Relative Method

When this method is used to construct a price index, price relatives are obtained for the various items
included in the index and then an average of these relatives is obtained using any one of the measures of central
tendency, i.e. arithmetic mean, median, mode, Geometric mean or harmonic mean. When arithmetic mean is used
for averaging the relatives, the formula for computing the index is:

P
∑ P1 ∗100
0
P 01=
N
Where N refers to number of items (commodities) whose price relatives are thus averaged.

Although any measure of central tendency can be used to obtain the overall index, price relatives are generally
averaged either by the arithmetic mean or the geometric mean. When geometric mean is used for averaging the
price relatives the formula for obtaining the index becomes
P
∑ log[ P1∗100 ]
log P01=
0
or
∑ log P
N N ;
P1
P= ∗100
Where P2

Illustration 2:
From the data of illustration 1, compute price index by simple average of price relatives method based on

Page 10 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

a. Arithmetic mean b. Geometric mean

Solution:
a. Price index based on simple average of price relatives

Price relative (
Commodity and Price (tk.) in 2007 Price (tk.) in 2008 P
Unit ( Po ) ( P1 ) P= 1∗100
P0 )
Butter (kg.) 110 120 109.09
Cheese (kg.) 75 80 106.67
Milk (lt.) 13 13 100.00
Bread (lb.) 9 9 100.00
Eggs (Doz.) 18 20 111.11
Ghee (1 Can) 850 860 101.18
P
N=6 ∑ P1∗100= 628.05
0
Price index

1 P
∑ ∗100
P is a net increase in price of commodities included in the index
This means that as compared to 2007, in 2008 there 0 628 . 05
=P01= = =104 . 67
to the extent of 4.67%. N 6

b. Price index based on geometric mean of price relatives

Price (tk.) in
Price (tk.) in Price relative ( log P
Commodity and
2008 ( P1 )
P
Unit ( Po ) P= 1∗100
2007 P0 )
Butter (kg.) 110 120 109.09 2.038
Cheese (kg.) 75 80 106.67 2.028
Milk (lt.) 13 13 100.00 2.000
Bread (lb.) 9 9 100.00 2.000
Eggs (Doz.) 18 20 111.11 2.046
Ghee (1 Can) 850 860 101.18 2.005
N=6 ∑ log P= 12.117

∑ log P =anti log 12. 117

P 01 =anti log
[ N ] [ 6 ] =anti log 2. 019=104 .579
4

This means that as compared to 2007, in 2008 there is a net increase in price of commodities included in the index
to the extent of 4.57%.

4
Calculating Antilog in EXCEL Spread sheet: The anti log (10 base) for a given value 2.019 is calculated by using
^ .019
¿2
the equation 10 ¿ .

Page 11 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

Note: Although arithmetic mean and geometric mean have both been used, the arithmetic mean is often preferred
because it is easier to compute and much better known.

Merits and limitations of the methods:

This method has the following two advantages over the previous method:
a. Extreme items do not influence the index. Equal importance is given to all the items.
b. The index is not influenced by the units in which prices are quoted or by absolute level of individual
prices. Relatives are pure numbers and are, therefore, independent of the original units. Consequently,
index numbers computed by the relative method would be the same regardless of the way in which prices
are quoted
Despite these merits this method is not very satisfactory because of the following reasons:
a. Difficulty is faced with regards to the selection of the appropriate average. The use of arithmetic mean is
considered as questionable sometimes because it has an upward bias. The use of geometric mean
involves difficulties of computations. Other averages are almost never used while constructing index
numbers.
b. The relatives are assumed to have equal importance. This is again a kind of concealed weighting system
that is highly objectionable since economically some relatives are more important than others.

ii. Weighted Index Number:

Weighted index numbers are of two types:
a. Weighted Aggregative Index numbers
b. Weighted average of relative Index numbers

These index numbers are of the simple aggregative type with the fundamental difference that weights are
assigned to the various items included in the index. There are various methods of assigning weights and
consequently a large number of formulae for constructing index numbers have been devised of which some of the
more important ones are:
i. Laspeyres Method,
ii. Paasche Method,
iii. Dorbish and Bowley’s method
iv. Fisher’s ideal method
v. Marshall – Edgeworth method
vi. Kelly’s method

The methods are discussed elaborately

i. Laspeyres Method:
In this method the base year quantities are taken as weights. The formula for constructing index is:
∑ P1 Q0
P 01 = ∗100
∑ P0 Q0

Page 12 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

Laspeyers index is very widely used in practical work and it attempts to answer the question: “What is the change
in aggregate value of the base period list of goods when valued at given period prices?”

ii. Paasche Method:

In this method the current year quantities are taken as weights. The formula for constructing index is:

P 01=
∑ P1 Q1 ∗100
∑ P0 Q 1
iii. Dorbish and Bowley’s Method:
Dorbish and Bowley have suggested simple arithmetic mean of the two indices (Laspeyres and Passche)as
mentioned above so as to take into account the influence of both the periods, i.e., current as well as base periods.
The formula for constructing the index is:
∑ P1 Q 0 + ∑ P1 Q 1
L+ P ∑ P 0 Q0 ∑ P0 Q1
P01 = = ∗100
2 2 ; [Where L= Lasperes Index and P= Paasche Index]
iv. Fisher’s ‘Ideal’ Method:
Prof. Irving Fisher has given a number of formulae for constructing index number and of these he calls one as the
‘ideal’ index. The Fisher’s Ideal index is given by the formula:

∑ P1 Q0 ∗∑ P1 Q1∗100
P01=√ L∗P=
√ ∑ P0 Q 0 ∑ P 0 Q 1 ; [Where L= Laspeyres Index and P= Paasche Index]
It shall be clear from the above formula that fisher’s ideal index is the geometric mean of the Laspeyres and
Paasche indices.

The above formula is known as ‘ideal’ because of the following reasons:

a) It is based on the geometric mean which is theoretically considered to be the best average for
constructing index numbers.
b) It takes into account both current year as well as base year prices and quantities.
c) It satisfies both the time reversal test as well as the factor reversal test as suggested by Fisher.
d) It is free from bias. The two formulae (Laspeyres and Paasche’s) that embody the opposing types and
weight biases are, in the ideal formula, crossed geometrically, i.e., by an averaging process that of itself
has no bias. The result is the complete cancellation of biases of the kinds revealed by time reversal and
factor reversal tests.

v. Marshall – Edgeworth Method:

In this method the current year as well as base year prices and quantities are considered. The formula for
constructing the index is:

P01=
∑ (Q0+ Q1) P1∗100= ∑ P1 Q0 +∑ P1 Q1 ∗100
∑ (Q0+Q 1) P0 ∑ P0 Q 0 + ∑ P 0 Q 1
vi. Kelly’s Method: T.L. Kelly Has suggested the following formula for constructing index number:

Page 13 of 15
LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

P01=
∑ P1 Q ∗100
∑ P0Q
Here weights are the quantities which may refer to some period, not necessarily the base year or current year.
Thus, the average quantity of two or more years may be used as weights .if in the Kelly’s formula the average of
the quantities of two years is used as weights, the formula becomes

P01=
∑ P1 Q ∗100 Q0 +Q 1
Q=
∑ P0Q ; [Where 2 ]
Similarly, the average of the quantities of three or more years can be used as weights. The method is known as
fixed weight aggregative index and is currently in great favor in the construction of index number series. An
important advantage of this formula is that like Laspeyres’ index it does not demand yearly changes in the
weights.

Illustration: Construct index numbers of price from the following data by applying:
i. Laspeyres Method,
ii. Paasche Method,
iii. Dorbish and Bowley’s method
iv. Fisher’s ideal method
v. Marshall – Edgeworth method

Commodities 2007 2008

Price Quantities Price Quantities
A 2 8 4 6
B 5 10 6 5
C 4 14 5 10
D 2 19 2 13

Solution: Calculation of various Indices

Commoditie 2007 2008

s P0 Q 0 P0 Q1 P1 Q0 P1 Q 1
Price ( Quantities ( Price ( Quantities (
 
P0 ) Q0 ) P1 ) Q1 )
A 2 8 4 6 16 12 32 24
B 5 10 6 5 50 25 60 30
C 4 14 5 10 56 40 70 50
D 2 19 2 13 38 26 38 26
Total 160 103 200 130

1. Laspeyres ∑ P1 Q0 200
P 01= ∗100 ∗100=125
Method: ∑ P0 Q0 = 160
2. Paasche’s
P 01=
∑ P1 Q1 ∗100=130 ∗100=126 . 21
103
Method: ∑ P0 Q 1

Page 14 of 15
 LS 05 STA 101 (SPRING 2015) I M SHAFIQUL KALAM (ISK)
Assistant Professor, MNS, BRACU

3. Bowley’s
∑ P1 Q 0 + ∑ P1 Q 1 200 130
+
L+ P ∑ P 0 Q0 ∑ P0 Q1 160 103
∗100=125. 605
Method: P01 = = ∗100
2 2 = 2
∑ P1 Q0 ∗∑ P1 Q1∗100
4. Fisher’s Ideal
Method:
P01=√ L∗P=
√
∑ P0 Q 0 ∑ P 0 Q 1 =
∑ (Q0+ Q1) P1∗100= ∑ P1 Q0 +∑ P1 Q1 ∗100
(
200 130
√ ∗ )∗100=125 .6
160 103

5. Marshall – P01=
∑ (Q0+Q 1) P0 ∑ P0 Q 0 + ∑ P 0 Q 1 =
Edgeworth
Method: 200+130
( )∗100=125. 475
160+103

Problem 1: Following table shows the prices and quantities of some of the vegetable items that Mirpur Zoo
authority usually buys to feed their herbivorous animals.
Price in Tk. Quantity
Item (Unit of quantity)
2005 2010 2005 2010
Cabbage (kg) 10 15 2000 1500
Carrot (Bunch) 20 30 200 200
Peas (kg) 20 40 400 500
Spinach (kg) 8 35 100 200
i. Compute Laspeyres price index for 2010 using 2005 as base period
ii. Compute Paasche’s price index for 2010 using 2005 as base period
iii. Compute Dorbish – Bowley’s price index for 2010 using 2005 as base period
iv. Determine Fisher’s ideal index for 2010 using 2005 as base period
Hence comment on the changes in the expenditure of the zoo authority to feed their herbivorous animal.

Question 2:
Following table shows the prices and quantities of some essential commodities for the year 2013 and 2014.
Price in Tk. Quantity
Items (Unit of quantity)
2013 2014 2013 2014
Rice (kg) 32 40 32.0 kg 25.0 kg
Oil (liter) 75 90 4.0 liter 3.0 liter
Lentil (Kg) 60 78 2.0 kg 1.5 kg
Milk (liter) 35 44 20 liter 15 liter
Meat (Kg) 200 240 10.0 kg 7.0 kg
Vegetable (Kg) 12 15 22.0 kg 20.0 kg
Using Fisher’s Ideal Index and Dorbish – Bowley’s methods - comment on the changes in the expenditure.

Page 15 of 15