Prepared By: Md.

Toufiq Hasan (24th Batch, AIS, RU)

Chapter 1
Time Series Analysis

#Population: A population is a complete set group of individuals, whether that

group comprises a nation or a group of people with a common characteristic.

#Types of Population
i) Finite Population: The finite population is also known as a countable

N
population, in which the population can be counted.
ii) Infinite Population: The infinite population is also known as an

SA
uncountable population, in which counting units in the population is not
possible.

#Variable: A variable is a quantity that varies from person to person or individual

HA
to individual.

#Types of Variables
i) Discrete Variables: Discrete variables are variables for which the values
they can take are countable and have a finite number of possibilities.
IQ
ii) Continuous variables: A continuous variable is defined as a variable that
can take an uncountable set of values or infinite set of values.
UF

#Time Series: A time series a set of numerical measurements on a time-dependent

variable of interest arranged over a regular interval of time. A time series is a
collection of observations made at specified times and arranged chronologically.
TO

The interval may be an hour, a day, a week, a month, or a calendar year. For
Example, Hourly temperature readings, daily sales in a shop, weekly sales in a
shop, weekly sales in a market, monthly production in an industry, yearly
agricultural production, and population growth in ten years are examples of time
series.
D

#Objectives of Time Series Analysis: The main objectives of analyzing the time
M

series data are to get a concrete idea about the past behavior of data so that
appropriate course of action for future can be taken. However, the objective can be
pointed out as follows:
1) To identify the pattern and trend, and isolate the influencing factors or
effects.
2) To apply the idea obtained from analyzing the pattern of time series data for
future planning and control.

1
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#Importance of Time Series Analysis in Business Decision-Making:

1. This is the most popular and, so far, the effective method for business
forecasting.
2. It helps in understanding the past behavior of economic process and in
predicting the future behavior.
3. It helps in planning future operations.
4. It helps in evaluating current achievement.
5. It facilitates comparison of same phenomenon over two or more periods.

N
# Mathematical Models for Time-Series Analysis: There are two mathematical
models, which used for the decomposition of a time series into different

SA
components. These are:
• Multiplicative Model
• Additive Model

HA
# Multiplicative Model: In traditional or classical time series analysis, it is
assumed that there is a multiplicative relationship among these four components.
Let Yt denotes the value of a series at time t. symbolically,
Yt = Tt × St × Ct × It
IQ
where, Tt = Secular Trend Component, St = Seasonal component, Ct = Cyclical
component and It = Irregular component.
UF

For example, if Tt =450, St = 1.4, Ct = 1.6 and It = 0.9, then, Yt = 450 × 1.4 × 1.6 ×
0.9 = 907.2

# Additive Model: According to this model, a time series is the sum of its four
TO

components. Symbolically,
Yt = Tt + St + Ct + It
where, Tt = Secular Trend Component, St = Seasonal component, Ct = Cyclical
component and It = Irregular component.
D

For example, if Tt = 250, St =45, Ct = 12 and It = -6, then Yt = 250 + 45 + 12 – 6 =

M

301

# Components of Time Series: The important components of a time series are as

follows -
• Secular Trend (Tt)
• Seasonal Trend (St)
• Cyclical Fluctuation (Ct)
• Irregular Variation (It)
2
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

# Secular Trend (Tt): Many time series met in practice exhibit a tendency of either
growing or reducing fairly steadily over time. This tendency of a time series data
over a long period of time is called secular trend. Some series increase slowly,
some increase fast, others decrease at varying rate, and some remain relatively
constant for long periods of time.
For example, despite short-run deviations from the trend, gross domestic product
of a country, prices or productions of a farm or a country, money supply show an

N
upward trend over the years due to increasing population. On the other hand,
population growth or death rate shows a downward trend. The time series data can
be represented by a Histogram, which exhibits its long-term trend.

SA
# Factors affecting trend: There are several factors that affect trend in time series
data. Some of the important factors are:

HA
1. Population: The ever-increasing population of a country is responsible for
increasing trends in series like prices, population, production sales etc.
2. Technology, Institution and culture: Downward or upward trends in some
factors are caused by technological, institutional or cultural changes.
IQ
# Reasons for studying Trends: The reasons for studying trend in a time series
data are pointed below:
1. It allows us to describe the historical pattern of time series data.
UF

2. It permits us to project past pattern or trends into future.

3. It facilitated us to eliminate the trend component from the series in order to
obtain the de-trended series that is useful for studying other components of
time series data.
TO

# Seasonal Variations (St): Seasonal variations are like cycles, but they occur
short and repetitive calendar periods. By seasonal variation we mean a periodic
movement that repeats itself with remarkable similarity at a regular interval of
D

time, the period being no longer than one year. Hence, seasonal component of a
time series data is the repetitive and predictable movement of observations around
M

the trend line during particular time intervals of the year. In order to measure or to
detect the seasonal component, the data must be given in small unit of time, such
as hours, days, weeks, months or quarters. Many business and economic time
series met in practice consists of quarterly or monthly observations.

# Factors affecting seasonal variations: Seasonal variations in a series are also

caused by some factors. Some of the important factors are:

3
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

(i) Climate and weather factors: Change in the climate and weather conditions such
as humidity, heat, rainfall etc. act on different product and industries differently
and cause change in demand of them.
(ii) Social customs and Religious factors: Due to some religious festival the sales
or demand of practical product varies over the year.

# Cyclical Fluctuation: A cycle is a wave like pattern about a long-term trend that
is usually apparent over a number a year. The term cycle refers to the recurrent

N
variations in time series that usually last longer than a year. One complete period is
called a cycle. Many business and economic time series met in practice appear to
exhibit oscillatory, or cyclical, pattern unconnected with special behavior. They are

SA
not necessarily regular, but follow rather smooth pattern of upswings or
downswings. Examples of cycles include the business cycle that record periods of
economic recession and inflation, long term product-demand cycles and cycles in

HA
the monetary and financial sectors. There are four well-defined periods or phases
in the business cycle, namely i) prosperity, ii) decline, iii) depression and iv)
improvement.

# Irregular variations: So far we have discussed three sources of variability in a

IQ
time series. If the components of a time series are found to be trend, seasonal and
cyclical, we would expect the series very smooth and rather easily projected
forward to produce forecast. But this may not be the case practice. Because of an
UF

inevitable presence of another component known as irregular element, induced by

the multitude of factors influencing the behavior of any actual series and whose
pattern becomes rather unpredictable on the basis of past experience.
TO

# There are two reasons for identifying the irregular components in time series
data. These are:

(i) To suggest that on occasions it may be possible to explain certain moments in

D

the data as due to specific causes and to simplify further analysis.

(ii) To emphasis the fact that prediction of economic conditions is always subject
M

to certain degree of error owing to the unpredictable erratic, influences, which may
enter.

#Measurement of Trend Component

Of the four components of time series, secular trend represents the long-term
direction. One way to describe the trend component is to fit a line visually to a set
of points on a graph. Trend may be linear and non-linear.

4
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

Q: What are the various method for measuring trend components?

A: The following are the methods used to eliminate linear trend component from a
given time series data:
i) Graphic or free hand curve method.
ii) Semi-average method.
iii) Least squares method.
iv) Moving average method.

N
i) Graphic method: The given data are plotted on a graph paper and a free hand
trend line fitted to the data is obtained just by inspection. A freehand curve drawn
through the data values is often an easy and perhaps, adequate representation of

SA
data. Forecasts can be obtained simply by extending the line.
A trend line fitted by this method should conform to the following conditions:
a) The trend line should be smooth - a straight line or mix of long gradual curves.

HA
b) The numerical sum of vertical deviations of the observations below the trend
should be equal to the numerical sum of deviations above the line.
c) The sum of squares of deviations of the observations from the trend line should
be as small as possible.
d) The trend line should bisect the cycles so that area above the trend line should
IQ
be equal to the area below the line, not only for the entire series but as much as
possible for each full cycle.
UF

Example: Fit a trend line to the following data by usig freehand curve method.

Year 1996 97 98 99 2000 01 02 03 04 05 06 07 08

TO

Sales
80 85 97 110 160 94 86 174 180 200 135 120 105
Turnover
Solution At first we plot the
turnover against year in a graph
D

paper. Then we draw a free hand

trend line on the graph, the
M

estimated trends are the values

lying on the line corresponding to
respective years. Forecast of the
turnover can be obtained simply
by extending the trend line.

5
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

# Merits and Demerits of the Freehand curve method

Åpparently this method is very simple to determine the trend values of a time
series data. However, this method has the following merits and demerits:

Merits
i) This is the simplest, quickest and easiest method of estimating trend values.
ii) This method is very flexible in the sense that it can be used irrespective of the
nature of the trend component, whether it is linear or non-linear.

N
iii) If the statistician knows the past behavior of the data series, it is possible to
obtain a secular trend by this method, even sometimes better than by any other
mathematical method.

SA
Demerits
i) From its nature of fitting, it is clear that this method is very subjective, because

HA
the trend line depends on personal judgment and therefore what happens to be a
good fit to one individual may not be so for others.
ii) The trend line drawn cannot have much value if it is used as a basis for
prediction.
iii) It is very time consuming to construct a free hand curve if a careful and
IQ
conscientious job is to be done.

ii) Semi-average method: In this method, the given data are divided into two
UF

equal parts preferably with equal number of periods. If there is an odd number of
years or period like 7, 9 etc. the middle year is ignored and the two equal parts are
formed. An average of each part is computed and the two points thus obtained are
centered corresponding to the middle period and shown on the graph. A straight
TO

line is drawn through these two points. The values lying on this line describe the
trends. By projecting the line it is possible to forecast the future values.

# Merits and Demerits of the Semi-average method

D

Like the free hand curve method, the semi-average method of determining the
trend has the following merits and demerits.
M

Merits
i) This is a more logical and easier method of determining trend than freehand
curve method;
ii) This method does not take that much time to find the trend component;
iii) This is an objective method of determining trends as everyone whoever applies
this method is bound to get the same result.

6
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

Demerits
i) This method assumes straight line relationship between the time period and the
observations, so if the relationship deviates much from the linearity, then forecast
by this method will be biased and less reliable;
ii) Since this method is based on arithmetic mean, it also bears the limitations of
this mean. For example, if there are some extreme values in either half or both half
of the series, then the trend line will not indicate the true picture of the growth
factor. This danger is the greatest if the time period represented by the average is

N
small.

Example: Apply the method of semi-average of measuring the trend of sales of

SA
commodity from the following data.

Year Sales (in ‘000)

HA
2010 20
2011 24
IQ
2012 22
2013 30
UF

2014 28
2015 32
TO

Solution: Since there are six years, we will take an average of the first three years
namely 2010, 2011, 2012 and last three years namely 2013, 2014, 2015. These
averages will be plotted corresponding to the middle period i.e. 2011 and 2014
respectively. When these two are joined we will get the trend line by the method of
D

semi-averages.
M

Now, average of first three observations is (20+24+22)/3 = 22 which is plotted

corresponding to 2011 and average of last three observations is (30+28+32)/3 = 30
which is plotted corresponding to 2014.

7
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

Trend values of sales obtained by semi-average method

Year Sales (in ‘000) Semi-total Semi-average Trend Values

2010 20 19.33
22 + 24 + 22 =
2011 24 22 22.00
66
2012 22 24.67

N
2013 30 27.33

SA
30 + 28 + 32 =
2014 28 30 30.00
90
2015 32 32.67

HA
Trend Value Calculation: Semi-average increment = (30 - 22) = 8/3
2010 : 22 - 8/3 = 19.33 ; 2012 : 22 + 8/3 = 24.67
2013 : 30 - 8/3 = 27.33 ; 2015 ; 30 + 8/3 = 32.67
2011 : 22
IQ
2014 : 30

The original sales figures and the estimated trend valuesare shown in following
UF

graph.
TO
D
M

8
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#Least squares method for fitting trend line:

Simple regression analysis: y = a + bx

Here, y is a dependent variable

x is a independent variable
a and b are constant
y = a + bx
or, Σ𝑦 = Σ(𝑎 + 𝑏𝑥)

N
or, Σ𝑦 = 𝑛𝑎 + 𝑏Σ𝑥 ………… (i)
2
or, Σ𝑥𝑦 = 𝑎Σ𝑥 + 𝑏Σ𝑥 ………. (ii)

SA
Σ𝑥.Σ𝑦
Σ𝑥𝑦 − 𝑛
𝑏= 2 (Σ𝑥)
2 𝑎 = 𝑦 − 𝑏𝑥
Σ𝑥 − 𝑛

HA
#Problem: The following table shows the production of a fertilizer factory.

Year Production (in ‘000 Tons)

IQ
2014 70
2015 75
UF

2016 92
2017 98
TO

2018 89
2019 91
2020 100
D

Requirement:
(i) Calculate the values for straight-line by the squares method.
M

(ii) Estimated production for 2023.

(iii) Estimate the trend value.

#Solution:
(i) y = a + bx
Σ𝑥.Σ𝑦
Σ𝑥𝑦 −
𝑏= 2
𝑛
(Σ𝑥)
2 , 𝑎 = 𝑦 − 𝑏𝑥
Σ𝑥 − 𝑛

9
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

2014 − 2017
Here, x = 1
= -3

Year y x xy x2
2014 70 -3 -210 9
2015 75 -2 -150 4

N
2016 92 -1 -92 1
2017 98 0 0 0

SA
2018 89 1 89 1
2019 91 2 182 4

HA
2020 100 3 300 9
Total 615 0 119 28
IQ
Σ𝑥.Σ𝑦
Σ𝑥𝑦 −
Here, 𝑏 = 2
𝑛
2
(Σ𝑥)
Σ𝑥 − 𝑛
UF

0×615
119 − 7
= 2
0
28 − 7
TO

119
= 28
= 4.25

𝑎 = 𝑦 − 𝑏𝑥
D

615
= − 4. 25 × 0 = 87.8571
M

The fitted line (regression line) trend line is y = 87.85 + 4.25x

87.85 + 4.25
= 113.35
6 🇽
(ii) The estimated production for 2023 is -

10
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

(iii) Estimated the trend value:

x Trend Line
Year y
87.85 + 4.25x
2014 70 -3 75.1
2015 75 -2 79.35

N
2016 92 -1 83.6

SA
2017 98 0 87.85
2018 89 1 92.1
2019 91

HA
2 96.35
2020 100 3 100.6
Total 615 615
IQ
UF
TO
D
M

11
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

Chapter 2
Probability Distribution

#Random Variable: A random variable is a variable that takes value from an

experiment. There are to types of random variables:
(i) Discrete random variable and
(ii) Continuous random variable

N
#Probability distribution: Probability distribution is related to theoretical
frequency distribution. The probability distribution describes how outcomes are

SA
expected to vary. If we toss a fair coin twice, the possible outcomes are - HH, HT,
TH, TT.
P(H) = 2,1,1,0 / {0,1,2}

HA
#Types of Probability Distributions:
1. Discrete Probability Distributions: Discrete random variable
a. Binomial Distribution
b. Poisson Distribution
2. Continuous Probability Distributions: Continuous random variable
IQ
a. Normal Distribution

#Discrete Probability Function P(x) & Continuous Probability Function f(x).

UF

#Binomial Distribution: The most frequently encountered discrete probability

distribution is the binomial probability distribution. def. A discrete random variable
TO

X is said to have a binomial distribution if its probability function is defined by -

𝑝(𝑥) 𝑜𝑟 𝑝(𝑋 = 𝑥) = nCx px qn-x ; x = 0, 1, 2,...., n

or, ( )p q
𝑛
𝑥
x n-x
D

#Find the mean and variance of binomial distribution.

M

If n and p are two parameters of the binomial distribution, then mean and variance
of the distribution are -

Mean = μ = E[X] = np and variance = σ2 = npq

Standard deviation = σ = 𝑛𝑝𝑞

It is to be noted that mean is greater than variance of the distribution since q < 1.

12
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#Give some practical applications of binomial distribution:

1. Number of defective items in a randomly selected sample of 12 products.
2. Number of days of increasing price index in the share market in a randomly
selected 15 days.
3. Number of workers suffering from occupational disease in a randomly
selected sample of 10 workers.
4. Number of female babies in society.
5. Number of customers buy a particular commodity.

N
6. Number of successful hits in a target out of fixed number of hits.

#What are the properties or characteristics of the binomial distribution.

SA
1. It is a discrete probability distribution with parameters n and p.
2. The mean of the distribution is np and its variance is npq.
3. The distribution is symmetric if p = q = 0.5

HA
4. The distribution tends to Poisson distribution if the number of trials, n tends
to infinity.
5. The distribution is positively skewed if p < 0.5 and negatively skewed if p >
0.5.
IQ
#Problem: It was found that 5% of products of a lot are defective. If 8 products are
selected randomly, What is the probability of getting less than 3 defective
products?
UF

Solution: Here, p = 5% = 0.05

q = 1 - 0.05 = 0.95
n=8
TO

We know that, P(x) = nCx px qn-x

We have to find out P(x < 3)

D

P(x < 3) = P(x = 0) + P(x = 1) + P(x = 2)

M

= 8C0 (0.05)0 (0.95)8-0 + 8C1 (0.05)1 (0.95)8-1 + 8C2 (0.05)2 (0.95)8-2

=1 × 1 × 0. 6634 + 8 × 0. 05 × 0. 6983 + 28 × 0. 0025 × 0. 7351
= 0.6634 + 0.2793 + 0.0515
= 0.9942
= 99.42%

13
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#Poisson Distribution: Another discrete probability distribution is Poisson

Distribution. A discrete random variable X is said to have a Poisson probability
distribution if its probability function is defined by -
−λ 𝑥
𝑒 λ
𝑃(𝑥) = 𝑥!
; x = 0, 1, 2,....., ∞ ; e = 2.71828 and λ is a parameter.

Mean and variance are same in Poisson distribution equal to λ.

#Some important properties of Poisson distribution:

N
1. The mean of the distribution is λ.
2. The variance of the distribution is also λ.

SA
3. When λ tends to ∞, Poisson distribution tends to normal distribution.
4. The distribution is positively skewed and leptokurtic.

HA
#Normal distribution: A random variable X is said to be have a normal
distribution if its probability density function is defined by -
1 2
− (𝑥−µ)
2 1 2

f(x ; μ ; σ ) = .𝑒 2σ
;-∞<x<+∞
σ 2π
Here π and e are mathematical constants. π = 3.1416 and e = 2.7183
IQ
The mean of normal distribution is μ and variance is σ2.
UF

#Some important properties of normal distribution:

1. The distribution is symmetrical about μ.
2. Mean, median and mode of normal distribution are always equal.
3. The mean of the distribution is μ and the variance is σ2.
TO

4. The curve has a single peak i.e. it is unimodal.

5. All odd central moments μ1, μ2, μ3 of the distribution are zero.
6. For large sample most of the distributions tend to normal distribution.
7. Skewness of the distribution is zero i.e. β1 = 0.
D

8. The distribution is mesokurtic and the value of β2 = 3 always normal

distribution.
M

#Standard normal distribution: A continuous random variable Z is said to be

have a standard normal distribution if it probability density function is defined by -
2
𝑧
1 −
f(z; 0 , 1) = .𝑒 2
;-∞<z<+∞
2π
The variable Z is called standard normal variate. The mean of the distribution is
zero and the variance is one. Z ~ N (0, 1)

14
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

Chapter 3
Sampling Distribution

#1. What is the sample?

Answer: The representative part of a population is called a sample.

#2. What is Population?

Answer: The totality of all units of a universe is called the population.

N
#3. What is statistic?

SA
Answer: The sample characteristics of a sample is called statistic. Or, Any
function of a random sample is known as statistic.

#4. What is a parameter?

HA
Answer: The population characteristic is called a parameter.

#5. What is sampling distribution?

Answer: The probability distribution of statistic calculated from all possible
random samples of a given population is called a sampling distribution.
IQ
The probability distribution of a statistic derived from all possible random samples
of a given population is called a sampling distribution.
UF

#6. Give some important Sampling distributions.

Answer: i) χ2 (Chi-Square) Distribution
ii) t-distributions
TO

iii) F-distributions
iv) Distribution of Sample mean
v) Distribution of Sample proportion

#7. What is the central limit theorem?

D

Answer: The central limit theorem states that ‘Regardless of the shape of the
population, the distribution of the sample means approaches the normal probability
M

distribution as the sample size increases.’

In practice, a sample size of 30 or more is considered adequate for this purpose.
However, regardless of the sample size, the sampling distribution would be normal,
if the original population is normally distributed.

15
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#8. Statement of Central Limit Theorem

Answer: Let, x1, x2, x3,......,xn be a random sample from a population having mean
μ and variance σ2. Let 𝑥 be the sample mean, the central limit theorem states that as
𝑋−µ
n becomes large, the sampling distribution of 𝑍 = σ approaches to the
𝑛

standard normal distribution.

𝑋−µ
That means, 𝑍 = ∼ N (0, 1)

N
σ
𝑛

SA
#9. What is Chi-Square (χ2) Distribution?
Answer: The sum of squares of n independent standard normal variates is called
chi-squares with n degrees of freedom. Let Z1, Z2, Z3, ... , Zn be n independent
standard normal variables, then chi-square denoted χ2 is defined as -

HA
𝑛
2 2 2
χ = ∑ 𝑍𝑖 = Σ 𝑍
n
𝑖=1

#10. Write down the important properties of Chi-Square distribution.

IQ
Answer: Important properties of χ2 distribution:
i) The distribution contains only one parameter, which is the degree of freedom (n)
of the distribution.
UF

ii) The mean of the distribution is n and the variance is 2n.

iii) The mode of the distribution is n-1.
iv) It is positively skewed distribution for smaller values of n; the distribution
becomes symmetrical as n tends to infinity. (n → ∞)
TO

#11. What is the sampling distribution of the sample mean?

Answer: Let X1, X2, X3,......,Xn be a random sample from a population with mean
𝑛
1
D

2
μ and variance σ . Then the sample mean is 𝑋 = 𝑛
∑ 𝑥𝑖
𝑖=1
M

#12. What is the mean of sample mean?

Answer: It can be easily shown that the expectation of the sample mean is equal to
𝑛
1
the population mean. That is 𝐸(𝑋) = 𝐸( 𝑛
∑ 𝑥𝑖 )
𝑖=1

1
= 𝑛
× 𝐸(𝑛𝑥) = 𝑥 or μ

16
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#13. What is the variance of sample mean?

Answer: We know that the variance of a linear combination of independent
random variables is the sum of linear coefficients squared times the variance of
random variables. It follows that
𝑉𝑎𝑟(𝑋) = 𝑉𝑎𝑟(
Σ𝑥
𝑛
) = 𝑉𝑎𝑟 ( 𝑥1 +𝑥2 +𝑥3 +.....+𝑥𝑛
𝑛 )
( 𝑥1 𝑥2 𝑥3 𝑥𝑛
)

N
= 𝑉𝑎𝑟 𝑛
+ 𝑛
+ 𝑛
+.... + 𝑛
2

SA
1 2 σ
= 2 × 𝑛σ = 𝑛
𝑛

#14. What is the standard error of sample mean?

HA
σ
Answer: The standard error of sample mean is .
𝑛

#15. Find the sample mean of 𝑥.

σ
IQ
Answer: 𝑥 ∼ (µ, )
𝑛

#16. What are the properties of sample mean?

UF

Answer: Sampling distribution of 𝑥 (sample mean) has μ and sampling distribution

σ
of 𝑥 has standard error .
𝑛
TO

Problem 1: Suppose light bulbs have a life length with an average of 4,000 hours
and a variance of 40,000. If we take a sample of 100 light bulbs and calculate
mean, what is the probability that mean will be: (i) between 3950 and 4100? (ii)
more than 4150?
D

2
Solution: (i) We have µ = 4000 , σ = 40000
𝑥−µ 3950− 4000 50
M

For 𝑥 = 3950 , z = σ = 200 = - 20 = -2.5 ≈ 2.5

𝑛 100

It represents an area of (0.9938-0.5) = 0.4938

𝑥−µ 4100− 4000 100

For 𝑥 = 4100 , z = σ = 200 = 20
= 5 (3.5 এর উপরে হলে সব সময় 0.5 ধরতে হবে)
𝑛 100

It represents an area of 0.5.

∴ Total area = 0.4938 + 0.5 = 0.9938 ≈ 99.38%
17
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

(ii) To find the probability that the sample mean will be more than 4150 hours, we
again need to calculate the z-score:
𝑥−µ 4150− 4000 150
For 𝑥 = 4150 , z = σ = 200 = 20
= 7.5
𝑛 100

It represents an area of 0.5.

∴ Total area = 0.5 - 0.5 = 0 ≈ 0%

N
#17. What is sampling distribution for proportion?

SA
Answer: Suppose, a variable has two categories which follows binomial
distribution with parameters n and π, and suppose a random sample of size n is
taken from the population, where P is the proportion of a particular category of the
variable. We know the mean and variance of distribution are nπ and nπ(1-π)

HA
respectively.
π (1−π) 𝑃−π
The standard error of estimated P is given by σ𝑝 = 𝑛
;z=
σ𝑝
Problem 2: A manufacturer of pens has determined from experience that 5% of the
IQ
pens he produces are defective. If a random sample of 420 pens is examined, what
is the probability that the proportion defective is between 0.024 and 0.044?
UF

π (1−π)
Solution: Here, σ𝑝 = 𝑛

0.05 (1−0.05)
= = 0.0106
TO

420

𝑃−π 0.024 − 0.05

For, 0.024, z = = 0.0106
= -2.45
σ𝑝
D

For z = -2.45 representing an area of 0.4929.

M

𝑃−π 0.044 − 0.05

For, 0.044, z = = 0.0106
= -0.57
σ𝑝

For z = -0.57 representing an area of 0.2157

The required probability is (0.4929 - 0.2157) = 0.2772 ≈ 27.72%

18
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

Problem 3: It is known that 65% of items of a lot are defective. What is the
probability that a simple random sample of size 100 items will show the proportion
of defective items to be 60% or less.

π (1−π) 0.65 (1−0.65) 0.65 × 0.35)

Solution: Here, σ𝑝 = 𝑛
= 100
= 100
= 0.048

𝑃−π 0.60 − 0.65

For, 0.60, z = = 0.048
= -1.04

N
σ𝑝

For z = -1.04 representing an area of 0.3508.

SA
The required probability is (0.5 - 0.3508) = 0.1492 ≈ 14.92%

HA
#17. What are the parts of statistic?
Answer: Statistic is mainly divided into two parts: (i) Statistical Inference (ii)
Statistical Method

#18. Define Estimator.

IQ
Answer: An estimator is a sample statistic used to estimate a population
parameter.
UF

#19. Define Estimate.

Answer: An estimate is a specific observed numerical value of a statistic used to
estimate a parameter.
TO

#20. Types of Estimation.

Answer: It is possible to make two types of estimations of a population parameter;
these are
(i) Point estimation and (ii) Interval estimation
D

#21. Define Point Estimate.

M

Answer: A point estimate is a single number that is used to estimate an unknown

population parameter.
For example: Observing the first quarter’s credit of a bank, the manager can say
that our average credit of certain year will reach to TK. 30 million; again,
observing the early first week’s sales of a retail shop, a shopkeeper can say that the
average daily sales of this month will be TK. 20,000.

19
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#22. What are the properties of a good estimator?

Answer: Desirable properties of a good estimator are-
(i) Unbiasedness (ii) Consistency (iii) Efficiency (iv) Sufficiency

#23. What is unbiasedness?

Answer: An estimator is said to be unbiased if its expected value is equal to the
parameter. For example, a sample mean 𝑋 is said to be an unbiased estimator of the
population mean μ if

N
𝐸(𝑋) = μ

SA
#24. What is consistency?
Answer: Consistency refers to the effect of sample size on the accuracy of the
estimator. A statistic is said to be a consistent estimator of the population parameter

HA
if it approaches the parameter as the sample size increases. Thus, sample mean 𝑋 is
said to consistent estimator of population mean μ if 𝑋 → μ as n → ∞.

#25. What is Efficiency?

Answer: An estimator is considered to be efficient, if its value remains stable from
IQ
sample to sample. For example, sample mean (𝑥) is the best estimator of
population mean (μ) than median or mode because variance of sample mean is less
than that of median or mode.
UF

#26. What is sufficiency?

Answer: An estimator is said to be sufficient if it uses all the information about the
population parameter contained in the sample. For example, the statistic sample
TO

mean uses all the sample values for its computation while mode and the median do
not, hence, sample mean is a better estimator in this sense.

#27. What is confidence interval estimator?

D

Answer: A confidence interval estimator for a population parameter is a rule for

determining a range or an interval based on sample information. For example, The
M

shopkeeper would say that it is very likely (with a confidence of 95%, say) that the
daily sales amount would be between Tk. 19500 and Tk. 20500.

#Case I. Confidence interval for population mean when a sample is drawn from a
normal population with known variance σ2.
For population mean μ is given , confidence level is 95%.
𝑋 ± 𝑧σ𝑥 = 𝑋 ± 1. 96σ𝑥

20
 Prepared By: Md. Toufiq Hasan (24th Batch, AIS, RU)

#Problem: A magazine editor wants to know the average family income of its
readers. A random sample of 49 readers yields income data with a mean of Tk.
13,000 and an estimate of the population standard deviation (σ) of Tk. 2,000.
Construct a 95% confidence interval estimate of the population mean.

Solution: Here 𝑋 = 13, 000, σ = 2, 000, 𝑛 = 49

σ 2000
σ𝑥 = = = 285. 714
𝑛 49

N
We have, 𝑋 ± 𝑧σ𝑥 = 13000 ± 1. 96 × 285. 714

SA
= 13000 ± 560

Thus the first value X1 = 13000 - 560 = 12440

and The second value X2 = 13000 + 560 = 13560

HA
The confidence interval is 12440 - 13560.
IQ
UF
TO
D
M

21