LAISHA
Open Laisha

GATE ST Statistical Inference I: Estimation: practice questions

GATE - Statistics (ST) · India · IITs and IISc on rotation (2027: IIT Madras)

Original practice questions written to the published syllabus · not past papers
10worked here, free
84written to this exam in this subject
11syllabus units

About GATE ST

GATE - Statistics (ST) is a postgraduate admission exam run by IITs and IISc on rotation (2027: IIT Madras), for candidates in India. Computer-based, 3 hours, 100 marks (15 General Aptitude + 85 subject). MCQ, MSQ and NAT; negative marking on MCQ only; on-screen virtual calculator.

Source: gate2026.iitg.ac.in.

Checked against the exam body's published information. Formats and fees change; confirm the current edition there before you book.

Statistical Inference I: Estimation in GATE ST

How to estimate well and know it: sufficiency and completeness, unbiasedness and minimum variance, the Cramer-Rao bound, the Rao-Blackwell and Lehmann-Scheffe theorems, the methods of moments, maximum likelihood and minimum chi-square, and confidence intervals.

Our 84 GATE ST questions in this subject: 32 easy, 38 medium, 14 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 GATE ST and are not shown on any other exam's page.

Question 1 · medium

For a sample of 10 from uniform (0, θ), both 2X̄ and (n + 1)X₍ₙ₎/n are unbiased for θ. What is the efficiency of 2X̄ relative to (n + 1)X₍ₙ₎/n?

  1. 0.25
  2. 1.00
  3. 4.00
  4. 0.30
Answer and working

A. 0.25

Var(2X̄) = 4 × θ²/12/n = θ²/30; Var((n + 1)X₍ₙ₎/n) = θ²/[n(n + 2)] = θ²/120. Efficiency of 2X̄ relative to the other = Var(other)/Var(2X̄) = 30/120 = 0.25. The inverse ratio 4.00 is the efficiency of the max-based estimator relative to 2X̄.

Question 2 · medium

Eight pump lifetimes are exponential with rate λ and total 40 months. What is the UMVUE estimate of λ?

  1. 0.175 per month
  2. 5.000 per month
  3. 0.225 per month
  4. 0.200 per month
Answer and working

A. 0.175 per month

ΣX ~ gamma(n, λ) with E(1/ΣX) = λ/(n − 1), so (n − 1)/ΣX = 7/40 = 0.175 is unbiased and a function of the complete sufficient statistic. The MLE 1/X̄ = 0.200 is biased upward; 5 months is the mean life, not the rate.

Question 3 · medium

In 10 Bernoulli trials there are 4 successes. What is the UMVUE estimate of p²?

  1. 0.1778
  2. 0.1333
  3. 0.1600
  4. 0.1200
Answer and working

B. 0.1333

E[T(T − 1)] = n(n − 1)p² for T binomial(n, p), so T(T − 1)/[n(n − 1)] = 12/90 = 0.1333. The MLE p̂² = 0.16 is biased upward.

Question 4 · hard

For Poisson counts, the Fisher information about λ per observation is 1/λ. What is the information per observation about η = log λ?

  1. λ
  2. 1/λ²
  3. 1/λ
  4. log λ
Answer and working

A. λ

I(η) = I(λ)(dλ/dη)² = (1/λ)(λ)² = λ. Information is not invariant to reparameterisation; on the log scale it grows with λ.

Question 5 · easy

Which statement about method-of-moments estimators is generally true?

  1. They always lie inside the parameter space, unlike MLEs
  2. They are consistent when the moments exist, but less efficient than the MLE in many models
  3. They are always functions of the minimal sufficient statistic
  4. They are always unbiased and attain the Cramér-Rao lower bound, like the MLE in large samples
Answer and working

B. They are consistent when the moments exist, but less efficient than the MLE in many models

By the law of large numbers and continuity, moment estimators are consistent; but they can be inefficient (2X̄ for the uniform), biased, not based on sufficient statistics, and outside the parameter space.

Question 6 · medium

Observations are iid uniform on (−θ, θ). The sample mean of the squared observations is 12. What is the method-of-moments estimate of θ?

  1. 6.928
  2. 36
  3. 3.464
  4. 6.0
Answer and working

D. 6.0

E(X) = 0 carries no information, so use E(X²) = θ²/3: θ̂ = √(3 × 12) = 6.0. √12 forgets the factor 3.

Question 7 · easy

Nine plot yields from N(μ, σ²) have Σ(xᵢ − x̄)² = 36. What are the MLE of σ² and the unbiased estimate s²?

  1. MLE 4.0; s² = 4.5
  2. MLE 4.5; s² = 4.0
  3. MLE 4.0; s² = 4.0
  4. MLE 2.0; s² = 2.12
Answer and working

A. MLE 4.0; s² = 4.5

MLE σ̂² = 36/9 = 4.0 (divisor n); s² = 36/8 = 4.5 (divisor n − 1). 2.0 is σ̂, the MLE of σ.

Question 8 · medium

Why is the Wilson score interval preferred to the Wald interval for a binomial proportion?

  1. It is exact, guaranteeing at least nominal coverage for every p and n
  2. It is always narrower than the Wald interval at the same confidence level and sample size
  3. It needs no normal approximation at all, unlike the Wald interval
  4. It inverts the score test, keeps coverage near nominal for small n and stays inside [0, 1]
Answer and working

D. It inverts the score test, keeps coverage near nominal for small n and stays inside [0, 1]

The Wilson interval uses the null standard error √(p(1 − p)/n) and solves for p, which corrects the Wald interval's undercoverage near 0 and 1. It is still an approximation and is not always narrower; exact (Clopper-Pearson) intervals guarantee coverage.

Question 9 · easy

How does Fisher's method of scoring differ from Newton-Raphson?

  1. It replaces the observed information −ℓ″(θ) by the expected information nI(θ) at each step
  2. It uses only the likelihood values, without any derivatives
  3. It replaces the log-likelihood by the residual sum of squares
  4. It replaces the score ℓ′(θ) by its expected value, which is zero at the true θ
Answer and working

A. It replaces the observed information −ℓ″(θ) by the expected information nI(θ) at each step

Scoring uses θ + ℓ′(θ)/[nI(θ)]. Expected information is positive, which stabilises the iteration. Using E(score) = 0 would never move; derivative-free methods are a different class.

Question 10 · medium

Which property does the EM algorithm have?

  1. It always reaches the global maximum of the likelihood, whatever starting value is chosen
  2. It gives standard errors directly at the final iteration
  3. It converges quadratically, faster than Newton-Raphson
  4. The likelihood never decreases from one iteration to the next, though convergence is linear and slow
Answer and working

D. The likelihood never decreases from one iteration to the next, though convergence is linear and slow

EM guarantees monotone increase of the likelihood but converges linearly and can stop at a local maximum; standard errors need extra computation (for example Louis's method).

A study plan for Statistical Inference I: Estimation in GATE ST

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 (40% of the 84 questions): Cramer-Rao inequality and Fisher information (9, 11%); Maximum likelihood: properties, invariance and asymptotic normality (9, 11%); Iterative maximum likelihood: Newton-Raphson and Fisher scoring (8, 10%); Method of moments (8, 10%).

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

  2. Then the middle band: Minimum chi-square and other methods (8, 10%); Minimum variance unbiased estimation; Rao-Blackwell and Lehmann-Scheffe theorems (8, 10%); Sufficiency, the factorisation theorem and minimal sufficiency (8, 10%); Interval estimation: pivotal quantities and coverage (7, 8%).

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

  3. Then the lighter units: Point estimation: unbiasedness, consistency and efficiency (7, 8%); Completeness, ancillarity and Basu's theorem (6, 7%); Large-sample confidence intervals (6, 7%).

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

  4. Leave time for the hard ones: Cramer-Rao inequality and Fisher information (3 hard of 9), Completeness, ancillarity and Basu's theorem (2 hard of 6), Maximum likelihood: properties, invariance and asymptotic normality (2 hard of 9).

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

  5. Bank the easy marks: Minimum chi-square and other methods (5 easy of 8); Point estimation: unbiasedness, consistency and efficiency (4 easy of 7).

    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 — GATE ST Statistical Inference I: Estimation

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.