LAISHA
Open Laisha

Indian Statistical Service Examination Probability Distributions & Sampling Distributions: practice questions

Indian Statistical Service Examination (with Indian Economic Service) · India · Union Public Service Commission

Original practice questions written to the published syllabus · not past papers
9worked here, free
32written to this exam in this subject
11syllabus units

About Indian Statistical Service Examination

Indian Statistical Service Examination (with Indian Economic Service) is a government recruitment exam run by Union Public Service Commission, for candidates in India. Written examination of six papers and a viva voce. Reported scheme: General English 100, General Studies 100, Statistics I and II (objective, 200 each), Statistics III and IV (descriptive, 200 each); viva 200.

Official site: upsc.gov.in.

We have not been able to confirm these details against the exam body's own specification, so fees, dates and pass marks are left out on purpose. Check the current format, eligibility, fees and dates on the official site.

Probability Distributions & Sampling Distributions in Indian Statistical Service Examination

Every standard distribution a statistician must recognise and derive from: discrete and continuous families, their relations and limiting forms, compound and truncated distributions, order statistics, the bivariate normal and the sampling distributions chi-square, t and F.

Our 32 Indian Statistical Service Examination questions in this subject: 8 easy, 19 medium, 5 hard, across 11 units.

9 practice questions, worked

Choose an answer, then open the working. Each explanation says why the right answer is right and why the others are not.

These 9 were written to Indian Statistical Service Examination and are not shown on any other exam's page.

Question 1 · easy

Which discrete distribution on the non-negative integers has the memoryless property P(X ≥ s + t | X ≥ s) = P(X ≥ t)?

  1. The Poisson distribution
  2. The negative binomial with r = 2
  3. The binomial distribution
  4. The geometric distribution
Answer and working

D. The geometric distribution

Only the geometric is memoryless among discrete distributions, just as the exponential is among continuous ones. The Poisson, despite its link with the Poisson process, is not memoryless as a count.

Question 2 · medium

X has a Cauchy distribution with location 5 and scale 2. What is its interquartile range?

  1. 4
  2. 1.349 × 2
  3. It does not exist, because no moments exist
  4. 2
Answer and working

A. 4

The Cauchy quartiles are θ ± σ = 3 and 7, so the IQR is 2σ = 4. The Cauchy has no mean or variance, but its quantiles exist. 1.349σ is the IQR of a normal.

Question 3 · medium

X is normal with standard deviation 2. What is its fourth central moment μ₄?

  1. 24
  2. 12
  3. 16
  4. 48
Answer and working

D. 48

μ₄ = 3σ⁴ = 3 × 16 = 48, so β₂ = μ₄/μ₂² = 3. 16 is σ⁴ alone; 12 is 3σ².

Question 4 · easy

What distinguishes a truncated sample from a censored sample?

  1. Truncated samples are biased towards the mean; censored samples have no bias at all
  2. Truncated samples arise only with discrete data; censored samples arise only with continuous data
  3. Truncated: values outside the range are neither seen nor counted; censored: they are counted but not measured
  4. Truncated: values outside the range are counted but not measured; censored: they are neither seen nor counted
Answer and working

C. Truncated: values outside the range are neither seen nor counted; censored: they are counted but not measured

In truncation the units beyond the limit never enter the data, so even their number is unknown. In censoring, the units are known to exist and to lie beyond the limit (a pump still running at the end of a test), but their exact values are not observed. Both occur with discrete and continuous data.

Question 5 · medium

In Pearson's system, the condition 2β₂ − 3β₁ − 6 = 0 (κ infinite) identifies which type?

  1. Type V, the inverse-gamma form
  2. Type I, the beta of the first kind
  3. Type VII, the Student-t form
  4. Type III, the gamma family
Answer and working

D. Type III, the gamma family

When 2β₂ − 3β₁ − 6 = 0 the denominator of κ vanishes and the solution is Type III, the gamma (including chi-square and exponential; the exponential has β₁ = 4, β₂ = 9, satisfying 18 − 12 − 6 = 0). Type V corresponds to κ = 1.

Question 6 · hard

A large sample of n = 100 plot yields is drawn from a normal population with σ = 10. What is the approximate standard error of the sample median?

  1. 0.1253
  2. 1.5708
  3. 1.0000
  4. 1.2533
Answer and working

D. 1.2533

Var(median) ≈ 1/[4nf(ξ)²] with f(ξ) = 1/(σ√(2π)), giving πσ²/(2n) = 1.5708; SE = 1.2533. 1.0000 is the standard error of the mean, σ/√n; 1.5708 is the variance, not the standard error.

Question 7 · medium

Which of these is a member of the exponential family?

  1. Cauchy with unknown location
  2. Shifted exponential with unknown location
  3. Weibull with unknown shape
  4. Weibull with known shape and unknown scale
Answer and working

D. Weibull with known shape and unknown scale

With known shape k, Y = Xᵏ is exponential, and the Weibull density has the form h(x)exp(−θxᵏ − A(θ)) with T = xᵏ. Unknown shape breaks the form; the Cauchy location and the shifted exponential (parameter-dependent support) are not members.

Question 8 · medium

The normal distribution N(μ, σ²) with σ² known is a one-parameter exponential family. What is its natural parameter?

  1. μ
  2. μ²/(2σ²)
  3. μ/σ²
  4. 1/σ²
Answer and working

C. μ/σ²

exp[−(x − μ)²/(2σ²)] = exp(−x²/(2σ²)) × exp(xμ/σ² − μ²/(2σ²)): the coefficient of T(x) = x is μ/σ². μ²/(2σ²) is A(θ), the normalising term.

Question 9 · hard

X has mean vector μ1 (all entries equal) and covariance σ²I with n = 10 and σ² = 4, not necessarily normal. Using E(X′AX) = trace(AΣ) + μ′Aμ with A = I − J/n, what is E[Σ(Xᵢ − X̄)²]?

  1. 36
  2. 3.6
  3. 4
  4. 40
Answer and working

A. 36

trace(AΣ) = σ² trace(A) = 4 × 9 = 36, and μ′Aμ = 0 because A annihilates a constant vector. Hence E(s²) = 36/9 = σ²: the divisor n − 1 makes s² unbiased. 40 uses trace n instead of n − 1.

A study plan for Probability Distributions & Sampling Distributions in Indian Statistical Service Examination

This plan follows where our questions actually fall, so the time goes where the marks are likely to be. It is our reading of the syllabus, not the exam body's weighting.

  1. First, the heavy units (50% of the 32 questions): Pearsonian system and fitting of distributions to agricultural data (6, 19%); Discrete distributions: Bernoulli, binomial, Poisson, geometric, negative binomial, hypergeometric and multinomial (4, 13%); Bivariate normal: marginal and conditional distributions (3, 9%); Continuous distributions: uniform, normal, exponential, gamma, beta, Cauchy, Weibull, lognormal (3, 9%).

    Read the manual chapters for these first and do every worked question you can find on them.

  2. Then the middle band: Distributions of quadratic forms in normal variables (Cochran's theorem) (3, 9%); Relations and limiting forms among distributions (3, 9%); Sampling distributions: chi-square, t and F and their properties (3, 9%); Compound, mixture and truncated distributions (2, 6%).

    Aim for one sitting per unit, then mixed questions across them.

  3. Then the lighter units: Exponential family of distributions (2, 6%); Non-central chi-square, t and F (2, 6%); Order statistics: distributions of single, joint and range (1, 3%).

    Learn the definitions, classifications and standard numbers; these units are more often recalled than worked.

  4. Leave time for the hard ones: Order statistics: distributions of single, joint and range (1 hard of 1), Non-central chi-square, t and F (1 hard of 2), Distributions of quadratic forms in normal variables (Cochran's theorem) (1 hard of 3).

    Come back to these after a first pass, untimed, and read every explanation, including for the options you rejected.

  5. Bank the easy marks: Bivariate normal: marginal and conditional distributions (2 easy of 3); Discrete distributions: Bernoulli, binomial, Poisson, geometric, negative binomial, hypergeometric and multinomial (2 easy of 4).

    Do these early in revision so they are fast on the day.

  6. Finish timed. Sit the subject as a timed set without notes, mark it, and send each wrong answer back to its unit above.

Last updated · How Laisha writes and checks its pages · Report a mistake

Ask Laisha about this

Opens Laisha's chat with your question ready to send. Sign in or start free.

Unlock every question — Indian Statistical Service Examination Probability Distributions & Sampling Distributions

US$10 a month · one price worldwide · cancel any time

  • Every practice question in this subject, with full working
  • Timed exam papers and progress by unit
  • Ask Laisha to explain any answer — up to 1,000 questions a month
Free or Smarter
FreeSmarter
Grow guides, pest pages, articlesRead allRead all
Ask Laisha5 a dayUp to 1,000 a month
Exam practiceThe sample on each pageEvery question, timed papers
Crop calendar and pest alerts—For your own fields
Calculators and tools—All
Uploads and reports—100 a month, real files
10 days free for every new account, no card.