STATISTICS NOTES (BA ECONOMICS HONS.
, DU - SEMESTER II)
UNIT 1: Sampling Distribution of a Statistic (12 Hours, 20
marks)
1.1 Concepts: - Statistic: A function of sample observations. E.g., sample
mean ( {X} ), sample variance ( s^2 ). - Parameter: A numerical
characteristic of a population. E.g., population mean ( ), variance ( ^2 ).
1.2 Sampling Distributions: - Probability distribution of a statistic based
on a random sample. - Example: If X is normally distributed, the sample
mean ( {X} N(, ^2/n) ).
1.3 Central Limit Theorem (CLT): - For a large sample size (n ( ) 30), the
sampling distribution of ( {X} ) is approximately normal, regardless of the
population distribution. - Example: Population with unknown shape, mean
( = 50 ), standard deviation ( = 10 ). Then ( {X} N(50, 10^2/n) ).
Important Concepts: Statistic vs Parameter, CLT, distribution of sample
mean and variance.
Example Questions: 1. Define a statistic and a parameter. Give two
examples of each. (5 marks) 2. Explain the Central Limit Theorem. Illustrate
it using an example. (5 marks) 3. A population has ( = 60 ) and ( = 12 ).
What is the probability that the sample mean of size 36 is greater than 62?
(10 marks)
Focus Topics for Exam: Central Limit Theorem, properties of sample mean
and variance, and practical application-based probability questions.
UNIT 2: Estimation (12 Hours, 30 marks)
2.1 Estimators and Estimation: - Point Estimation: Single value
estimate of a parameter. - Interval Estimation: Range of values with a
confidence level.
2.2 Methods of Estimation: - Method of Moments (MoM): Equates
sample moments to population moments. - Example: Estimate ( ) using
sample mean ( {X} ). - Method of Maximum Likelihood (MLE):
Maximizes the likelihood function. - Example: For Bernoulli(p), MLE of p is
sample proportion ( ).
2.3 Properties of Good Estimators: - Unbiasedness: E(( )) = ( ) -
Consistency: ( _n ) in probability as ( n ) - Efficiency: Minimum variance
among unbiased estimators. - Sufficiency: Contains all info about the
parameter.
2.4 Confidence Intervals: - Based on Z, t, ( ^2 ), F distributions. -
Example (Mean, known ( )): ( {X} Z_{/2} ) - Example (Mean, unknown
( )): ( {X} t_{n-1, /2} )
Important Concepts: Estimator definitions and properties, interval
estimation, Z and t confidence intervals.
Example Questions: 1. Compare and contrast MoM and MLE with
examples. (10 marks) 2. Define and explain unbiasedness, consistency, and
efficiency with suitable examples. (10 marks) 3. Construct a 95% confidence
interval for a sample mean of 120, sample standard deviation of 15, and
sample size of 25. (10 marks)
Focus Topics for Exam: Confidence intervals, properties of estimators,
estimation techniques (MoM and MLE), formula-based applications.
UNIT 3: Inference (9 Hours)
3.1 Statistical Hypothesis: - Assumption about population parameter. -
Null Hypothesis (H0): No effect or status quo. - Alternative Hypothesis
(H1): Statement we want to test.
3.2 Types of Errors: - Type I Error: Rejecting H0 when it is true (( )). -
Type II Error: Not rejecting H0 when it is false (( )). - Power of a Test: ( 1 -
); probability of correctly rejecting H0.
Example Questions: 1. Define Type I and Type II errors with illustrations.
Explain the concept of power of a test. (10 marks) 2. Distinguish between
null and alternative hypotheses. Provide suitable examples. (5 marks)
Focus Topics for Exam: Types of errors, power of test, formulating
hypotheses.
UNIT 4: Hypothesis Testing (12 Hours)
4.1 One-Sample Tests: - Z-test (known ( )) for mean: - ( Z = ) - t-test
(unknown ( )): - ( t = ) - Proportion Test: - ( Z = )
4.2 P-value: - Probability of observing a test statistic as extreme as the
sample value under H0. - If P-value ( < ), reject H0.
4.3 Two-Sample Tests: - Equality of Means (Independent Samples): -
( t = ) - where ( s_p^2 ) is pooled variance. - Equality of Variances (F-
test): - ( F = s_1^2 / s_2^2 ), compare with F-distribution.
Example Questions: 1. A sample of size 50 has a mean weight of 70 kg and
population ( = 5 ). Test if the mean weight differs from 68 kg at 5%
significance level. (10 marks) 2. Explain how to use a two-sample t-test.
Under what assumptions can it be used? (10 marks) 3. A manufacturer
claims average product life is 500 hours. A random sample of 20 shows a
mean of 480 hours and std deviation of 30. Test the claim using a t-test. (10
marks)
Focus Topics for Exam: One and two-sample tests, P-value interpretation,
hypothesis setup, application of Z, t, F, and chi-squared tests.
Overall Exam Strategy: - UNIT 1 (20 marks): Understand sampling
distributions, focus on derivations and standard problems from CLT. - UNIT 2
(30 marks): Focus heavily on estimator properties, estimation methods,
confidence intervals. - UNIT 3 & 4 (40 marks): Practice formulating
hypotheses and full applications of one/two-sample tests. Be ready to
calculate and interpret test statistics and p-values.
End of Notes
