LAISHA
Open Laisha

Indian Statistical Service Examination Statistical Inference II: Testing of Hypotheses & Sequential Analysis: 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
10worked here, free
27written 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.

Statistical Inference II: Testing of Hypotheses & Sequential Analysis in Indian Statistical Service Examination

The logic and optimality of tests: errors and power, the Neyman-Pearson lemma, uniformly most powerful and unbiased tests, likelihood-ratio tests and their large-sample forms, and Wald's sequential probability ratio test.

Our 27 Indian Statistical Service Examination questions in this subject: 6 easy, 15 medium, 6 hard, across 11 units.

10 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 10 were written to Indian Statistical Service Examination and are not shown on any other exam's page.

Question 1 · easy

What is the power of a test at a particular alternative θ₁?

  1. The probability of accepting H₀ when θ₁ is the true value
  2. The probability of rejecting H₀ when θ₁ is the true value
  3. The probability that H₁ is true given the observed data
  4. The probability of rejecting H₀ when H₀ is in fact true
Answer and working

B. The probability of rejecting H₀ when θ₁ is the true value

Power = 1 − β(θ₁) = P_θ₁(reject H₀). Accepting H₀ under θ₁ is the Type II error probability β; rejecting a true H₀ has probability α. The posterior probability of H₁ is a Bayesian quantity, not power.

Question 2 · easy

For a fixed sample size, the critical region of a test is made smaller. What happens to the error probabilities?

  1. α and β both fall
  2. α rises and β falls
  3. α and β both rise
  4. α falls and β rises
Answer and working

D. α falls and β rises

A smaller critical region rejects H₀ less often under every θ, so the Type I error probability falls and the Type II error probability rises. Lowering both needs more data or less noise.

Question 3 · medium

X₁, ..., Xₙ are iid N(μ, 1). For H₀: μ = 0 against the two-sided H₁: μ ≠ 0, which statement is correct?

  1. The two-sided z-test is UMP among all level-α tests
  2. A UMP test exists, but it needs randomisation
  3. No UMP test exists; the two-sided z-test is UMP among unbiased tests
  4. The one-sided test rejecting for large x̄ is UMP against both sides
Answer and working

C. No UMP test exists; the two-sided z-test is UMP among unbiased tests

The MP test against μ > 0 rejects for large x̄ and against μ < 0 for small x̄; no single test is best on both sides. The test |Z| > z_(α/2) is UMPU.

Question 4 · hard

Lifetimes are exponential with mean θ. With n = 5 and χ²₁₀ upper 5% point 18.307, the UMP size-0.05 test of H₀: θ ≤ 10 against θ > 10 rejects when ΣXᵢ exceeds what value?

  1. 91.535
  2. 183.07
  3. 86.78
  4. 0.9153
Answer and working

A. 91.535

2ΣXᵢ/θ ~ χ²₂ₙ = χ²₁₀, so reject when 2ΣXᵢ/10 > 18.307, i.e. ΣXᵢ > 10 × 18.307/2 = 91.535. Forgetting the factor 2 gives 183.07; the normal approximation gives 86.78.

Question 5 · hard

For H₀: θ = θ₀ against θ ≠ θ₀ in a one-parameter exponential family with statistic T, the UMPU test rejects when T < c₁ or T > c₂. Which conditions fix c₁ and c₂?

  1. Size α at θ₀, and equal tail areas α/2 in all cases
  2. Size α at θ₀, and zero derivative of the power function at θ₀
  3. Equal distances of c₁ and c₂ from the mean of T under θ₀
  4. Power equal to 1 − α at the two nearest alternatives
Answer and working

B. Size α at θ₀, and zero derivative of the power function at θ₀

Unbiasedness with a differentiable power function forces a minimum at θ₀: π(θ₀) = α and π′(θ₀) = 0, equivalently E_θ₀[Tφ(T)] = αE_θ₀(T). Equal tails result only when T is symmetric under θ₀.

Question 6 · hard

Ten Poisson counts total 30. What is −2 log Λ for testing H₀: λ = 2 against λ ≠ 2?

  1. 4.328
  2. 12.164
  3. 5.000
  4. 2.164
Answer and working

A. 4.328

λ̂ = 3. −2 log Λ = 2[ΣX log(λ̂/λ₀) − n(λ̂ − λ₀)] = 2[30 ln 1.5 − 10] = 2[12.164 − 10] = 4.328, compared with χ²₁. 5.000 is the score statistic; 2.164 forgets the factor 2.

Question 7 · medium

A Wald test of H₀: β₂ = β₃ = 0 in a model with 5 coefficients is carried out. What is its asymptotic null distribution?

  1. χ² with 5 degrees of freedom
  2. Standard normal
  3. χ² with 2 degrees of freedom
  4. χ² with 3 degrees of freedom
Answer and working

C. χ² with 2 degrees of freedom

The multiparameter Wald statistic (β̂ − 0)′[Cov(β̂)]⁻¹(β̂ − 0) for the 2 restricted coefficients is χ² with df equal to the number of restrictions, 2.

Question 8 · medium

A 2 × 2 table of adoption (yes/no) by training (yes/no) has cells a = 30, b = 20, c = 10, d = 40. What is Pearson's χ² without continuity correction?

  1. 16.67
  2. 4.08
  3. 15.04
  4. 0.17
Answer and working

A. 16.67

χ² = n(ad − bc)²/[(a + b)(c + d)(a + c)(b + d)] = 100 × 1000²/(50 × 50 × 40 × 60) = 16.67, with 1 df. With Yates' correction it is 15.04; omitting n gives 0.17.

Question 9 · easy

A sample of 11 plots gives s² = 6. What is the χ² statistic for H₀: σ² = 4?

  1. 16.5
  2. 1.5
  3. 13.5
  4. 15.0
Answer and working

D. 15.0

χ² = (n − 1)s²/σ₀² = 10 × 6/4 = 15.0, with 10 df. 16.5 uses n = 11 instead of n − 1; 1.5 is the ratio s²/σ₀² alone.

Question 10 · easy

In a contingency table with grand total 200, a cell's row total is 60 and its column total is 40. What is its expected frequency under independence?

  1. 24
  2. 50
  3. 30
  4. 12
Answer and working

D. 12

E = row total × column total/grand total = 60 × 40/200 = 12.

A study plan for Statistical Inference II: Testing of Hypotheses & Sequential Analysis 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 (52% of the 27 questions): Monotone likelihood ratio and uniformly most powerful tests (4, 15%); Tests based on Z, t, chi-square and F distributions (4, 15%); Hypotheses, the two types of error, size and power (3, 11%); Unbiased tests and UMPU tests for exponential families (3, 11%).

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

  2. Then the middle band: Likelihood ratio tests and Wilks' theorem (2, 7%); Neyman-Pearson lemma and most powerful tests (2, 7%); Power and sample-size calculation for common tests (2, 7%); Relation between tests and confidence intervals (2, 7%).

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

  3. Then the lighter units: Sequential probability ratio test: OC and ASN functions (2, 7%); Wald and score (Rao) tests (2, 7%); Multiple testing: family-wise error rate and false discovery rate (1, 4%).

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

  4. Leave time for the hard ones: Monotone likelihood ratio and uniformly most powerful tests (2 hard of 4), Likelihood ratio tests and Wilks' theorem (1 hard of 2), Power and sample-size calculation for common tests (1 hard of 2).

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

  5. Bank the easy marks: Tests based on Z, t, chi-square and F distributions (3 easy of 4); Hypotheses, the two types of error, size and power (2 easy of 3).

    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 Statistical Inference II: Testing of Hypotheses & S

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.