LAISHA
Open Laisha

ICAR AICE-JRF/SRF Nonparametric & Rank-Based Methods: practice questions

ICAR AICE-JRF/SRF - Agricultural Statistics, Computer Application, Bioinformatics (PhD admission and fellowship) · India · National Testing Agency for ICAR

Original practice questions written to the published syllabus · not past papers
10worked here, free
60written to this exam in this subject
11syllabus units

About ICAR AICE-JRF/SRF

ICAR AICE-JRF/SRF - Agricultural Statistics, Computer Application, Bioinformatics (PhD admission and fellowship) is a PhD, fellowship and lectureship exam run by National Testing Agency for ICAR, for candidates in India. Computer-based test, 2 hours, 120 MCQs, +4 / -1 (Part A general, Part B core group, Part C subject). 2026 exam held with AIEEA-PG in July 2026.

Source: icar.org.in.

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

Nonparametric & Rank-Based Methods in ICAR AICE-JRF/SRF

Run, sign, Wilcoxon, Mann-Whitney, Kruskal-Wallis, Friedman and Cochran's Q tests by name: when each applies, the statistic, null mean and variance, and degrees of freedom of the chi-square approximation.

Methods that hold without a normal model: the empirical distribution, goodness-of-fit, sign and rank tests for one, two and many samples, rank correlation, and tests used for scored or ranked agricultural data such as disease ratings and sensory panels.

Our 60 ICAR AICE-JRF/SRF questions in this subject: 11 easy, 24 medium, 25 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 ICAR AICE-JRF/SRF and are not shown on any other exam's page.

Question 1 · hard

Using the Dvoretzky-Kiefer-Wolfowitz inequality P(sup|Fₙ − F| > ε) ≤ 2exp(−2nε²), what half-width ε gives a 95% confidence band for F from n = 100 observations?

  1. 0.098
  2. 0.122
  3. 0.136
  4. 0.0136
Answer and working

C. 0.136

Set 2exp(−200ε²) = 0.05: ε = √(log 40/200) = 0.136. 0.098 is a pointwise interval at the median, not a band; omitting the factor 2 gives 0.122.

Question 2 · hard

Why does the distribution of sup|Fₙ(x) − F(x)| not depend on F when F is continuous?

  1. The central limit theorem (CLT) makes every EDF approximately normal
  2. F(Xᵢ) is uniform, reducing it to the uniform case
  3. The supremum is attained at the median, whose distribution is free of F
  4. Continuous distributions all have the same variance after standardisation
Answer and working

B. F(Xᵢ) is uniform, reducing it to the uniform case

By the probability integral transform, Uᵢ = F(Xᵢ) are iid uniform, and Fₙ(x) − F(x) evaluated over x equals the uniform EDF minus the identity over u ∈ (0, 1). So the supremum has the same distribution for every continuous F.

Question 3 · medium

Which goodness-of-fit statistic gives more weight to discrepancies in the tails of the distribution than the KS statistic does?

  1. The runs statistic
  2. The Anderson-Darling statistic
  3. Kendall's tau
  4. The sign statistic
Answer and working

B. The Anderson-Darling statistic

Anderson-Darling weights squared deviations by 1/[F₀(1 − F₀)], which is large near 0 and 1, making it sensitive in the tails, where KS is weakest.

Question 4 · hard

Five rainfall amounts (mm), 2.0, 5.5, 9.0, 14.0 and 25.0, are tested against a fully specified exponential distribution with mean 10 mm. What is the KS statistic D₅?

  1. 0.2231
  2. 0.9179
  3. 0.0821
  4. 0.1231
Answer and working

A. 0.2231

F₀(x) = 1 − e^(−x/10) gives 0.1813, 0.4231, 0.5934, 0.7534, 0.9179. D⁺ = max(i/n − F₀) = 0.0821; D⁻ = max(F₀ − (i − 1)/n) = 0.2231; D₅ = 0.2231. Checking only i/n gives 0.0821 and misses the gap just before each jump; 0.9179 is merely F₀ at the largest observation.

Question 5 · medium

Four varieties are ranked within each of 6 blocks; the rank sums are 8, 14, 17 and 21. What is Friedman's statistic?

  1. 27.0
  2. 99.0
  3. 9.0
  4. 3.0
Answer and working

C. 9.0

χ²_F = [12/(bk(k + 1))]ΣRⱼ² − 3b(k + 1) = 0.1 × 990 − 90 = 9.0, with 3 df; it exceeds 7.815. Forgetting to subtract 3b(k + 1) gives 99.0; subtracting 3bk instead gives 27.0.

Question 6 · hard

Under normal errors, what is the asymptotic efficiency of Friedman's test relative to the RCBD F-test for k = 4 treatments?

  1. 0.764
  2. 0.955
  3. 0.637
  4. 1.194
Answer and working

A. 0.764

ARE = 0.955k/(k + 1) = (3/π)(4/5) = 0.764. It approaches 3/π only as k grows; for k = 2 it is 2/π (the sign test).

Question 7 · medium

For binary responses in blocks with additional covariates and a random block effect, which model-based method is the usual alternative to Cochran's Q?

  1. A Poisson regression with identity link
  2. A Kruskal-Wallis test on the block totals
  3. A logistic mixed model
  4. A one-way ANOVA on the 0/1 values
Answer and working

C. A logistic mixed model

A generalised linear mixed model with logit link and random block effects handles covariates and unbalanced data and gives odds-ratio estimates; GEE gives population-average effects.

Question 8 · easy

What is the asymptotic relative efficiency of the Wilcoxon signed-rank test relative to the t-test when the data are normal?

  1. 0.955
  2. 0.864
  3. 1.000
  4. 0.637
Answer and working

A. 0.955

3/π ≈ 0.955. 2/π is the sign test's ARE; 0.864 is the Hodges-Lehmann lower bound over all continuous symmetric distributions.

Question 9 · medium

What is the asymptotic relative efficiency of the sign test relative to the t-test for Laplace (double exponential) data?

  1. 0.637
  2. 0.333
  3. 1.5
  4. 2
Answer and working

D. 2

For the Laplace distribution the sign test is twice as efficient as the t-test. 1.5 is the Wilcoxon/t ARE for the Laplace; 1/3 is the sign/t ARE for the uniform.

Question 10 · hard

How is the Mann-Whitney test related to the proportional-odds model for comparing two groups on an ordinal scale?

  1. It is closely related to the score test for the group effect in that model
  2. It is unrelated, because the model is parametric and the test is not
  3. It is identical to the Wald test for the cut-point parameters θⱼ
  4. It is the likelihood-ratio test of the proportional-odds assumption itself
Answer and working

A. It is closely related to the score test for the group effect in that model

For two groups, the score test of β = 0 in the cumulative logit model is essentially the Wilcoxon rank-sum test with mid-ranks, which is why the two usually agree; the model adds odds-ratio estimates and covariates.

A study plan for Nonparametric & Rank-Based Methods in ICAR AICE-JRF/SRF

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 (43% of the 60 questions): Kruskal-Wallis test (7, 12%); Sign test and Wilcoxon signed-rank test (7, 12%); Empirical distribution function and its properties (6, 10%); Friedman's test for randomised blocks (6, 10%).

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

  2. Then the middle band: Spearman's and Kendall's rank correlations (6, 10%); Asymptotic relative efficiency (5, 8%); Cochran's Q test (5, 8%); Mann-Whitney U test and two-sample Kolmogorov-Smirnov test (5, 8%).

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

  3. Then the lighter units: Run test for randomness (5, 8%); Analysing ordinal scores: disease ratings and sensory panels (4, 7%); Goodness of fit: chi-square and Kolmogorov-Smirnov tests (4, 7%).

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

  4. Leave time for the hard ones: Empirical distribution function and its properties (4 hard of 6), Asymptotic relative efficiency (3 hard of 5), Analysing ordinal scores: disease ratings and sensory panels (2 hard of 4).

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

  5. Bank the easy marks: Sign test and Wilcoxon signed-rank test (2 easy of 7); Goodness of fit: chi-square and Kolmogorov-Smirnov tests (1 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 — ICAR AICE-JRF/SRF Nonparametric & Rank-Based Methods

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.