GATE - Statistics (ST) · India · IITs and IISc on rotation (2027: IIT Madras)
Original practice questions written to the published syllabus · not past papersGATE - 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.
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 61 GATE ST questions in this subject: 18 easy, 26 medium, 17 hard, across 11 units.
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.
X is the number of failures before the 3rd success in independent trials with success probability 0.5. What is Var(X)?
A. 6
Negative binomial (failures version): Var = rq/p² = 3 × 0.5/0.25 = 6. 3 is the mean rq/p; 0.75 is the binomial-style rpq.
Failure times of a tractor part fit a Weibull distribution with shape parameter k = 0.7. What does this say about the hazard rate?
B. It decreases with age, pointing to early-life infant failures
Weibull hazard ∝ x^(k−1): decreasing for k < 1, constant for k = 1 (exponential), increasing for k > 1. A single Weibull cannot produce a bathtub curve; that needs a mixture or a different model.
Which of these relations is FALSE?
D. If T has Student's t with n df, T² has the F distribution with (n, 1) df
T = Z/√(V/n) gives T² = (Z²/1)/(V/n), which is F(1, n), not F(n, 1): the numerator has one degree of freedom. The other three relations are correct.
X₁, ..., X₄ are independent exponential variables with rate 2. What is the distribution of their sum?
A. Gamma with shape 4 and rate 2, mean 2
The MGF of the sum is [2/(2 − t)]⁴, the gamma (Erlang) MGF with shape 4 and rate 2: mean 4/2 = 2. Rate 8 is the distribution of the minimum of the four, not the sum.
X is Poisson with mean λ and Y is gamma with shape k (a positive integer) and rate 1. Which identity holds?
C. P(X ≥ k) = P(Y ≤ λ)
In a unit-rate Poisson process the k-th arrival time is gamma(k, 1), and at least k arrivals occur by time λ exactly when the k-th arrival is at or before λ: P(X ≥ k) = P(Y ≤ λ). A continuous Y has P(Y = λ) = 0.
X₁, ..., X₄ are iid uniform on (0, 1). What is the expected sample range?
A. 0.60
E(X₍₄₎) − E(X₍₁₎) = 4/5 − 1/5 = 3/5 = 0.60, matching R ~ beta(n − 1, 2) with mean (n − 1)/(n + 1). 0.75 uses (n − 1)/n.
For a random sample from a distribution with finite variance, X̄ and s² are independent. What does this property imply about the population?
B. The population is normal; the property characterises the normal family
Independence of the sample mean and sample variance holds if and only if the population is normal (Geary's characterisation; Lukacs). For non-normal populations Cov(X̄, s²) = μ₃/n, which is non-zero for skewed distributions, and dependence remains even for symmetric non-normal ones.
Which of these families is NOT a one-parameter exponential family?
A. Uniform on (0, θ), θ unknown
The support (0, θ) of the uniform depends on θ, which the exponential-family form forbids. Poisson (η = log λ), binomial with known n (η = logit p) and normal with known variance (η = μ/σ²) are all members.
In natural form the Poisson has f(x) = (1/x!) exp(ηx − A(η)) with A(η) = e^η. What do A′(η) and A″(η) give?
D. Both equal λ: A′(η) = E(X) and A″(η) = Var(X)
In a natural exponential family E(T) = A′(η) and Var(T) = A″(η). With A = e^η = λ, both are λ: the Poisson mean equals its variance.
For a random sample from a beta(a, b) distribution with both parameters unknown, what is the natural sufficient statistic?
A. (Σ log Xᵢ, Σ log(1 − Xᵢ))
log f = (a − 1) log x + (b − 1) log(1 − x) − log B(a, b), so T = (log x, log(1 − x)). Note ΣXᵢ and Σ(1 − Xᵢ) carry the same information and are not sufficient here. ΠXᵢ alone covers only a.
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.
Read the manual chapters for these first and do every worked question you can find on them.
Aim for one sitting per unit, then mixed questions across them.
Learn the definitions, classifications and standard numbers; these units are more often recalled than worked.
Come back to these after a first pass, untimed, and read every explanation, including for the options you rejected.
Do these early in revision so they are fast on the day.
Last updated · How Laisha writes and checks its pages · Report a mistake
US$10 a month · one price worldwide · cancel any time
| Free | Smarter | |
|---|---|---|
| Grow guides, pest pages, articles | Read all | Read all |
| Ask Laisha | 5 a day | Up to 1,000 a month |
| Exam practice | The sample on each page | Every question, timed papers |
| Crop calendar and pest alerts | — | For your own fields |
| Calculators and tools | — | All |
| Uploads and reports | — | 100 a month, real files |
An AI operating system for agriculture: free guides, manuals and tools written for every country.