LAISHA
Open Laisha

Indian Statistical Institute admission tests Nonparametric & Rank-Based Methods: practice questions

Indian Statistical Institute admission tests (BStat, MStat, JRF in Statistics) · India · Indian Statistical Institute

Original practice questions written to the published syllabus · not past papers
10worked here, free
56written to this exam in this subject
11syllabus units

About Indian Statistical Institute admission tests

Indian Statistical Institute admission tests (BStat, MStat, JRF in Statistics) is a postgraduate admission exam run by Indian Statistical Institute, for candidates in India. Written tests in two parts: an objective mathematics/statistics test and a descriptive subject test, followed by interview for some programmes (reported). JRF selection by written test and interview.

Official site: isical.ac.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.

Nonparametric & Rank-Based Methods in Indian Statistical Institute admission tests

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 56 Indian Statistical Institute admission tests questions in this subject: 20 easy, 27 medium, 9 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 Institute admission tests and are not shown on any other exam's page.

Question 1 · easy

What are E[Fₙ(x)] and Var[Fₙ(x)] for the empirical distribution function at a fixed x?

  1. F(x) and F(x)[1 − F(x)]
  2. F(x) and F(x)[1 − F(x)]/n
  3. 0.5 and 1/(4n)
  4. F(x)/n and F(x)[1 − F(x)]/n²
Answer and working

B. F(x) and F(x)[1 − F(x)]/n

Since nFₙ(x) ~ binomial(n, F(x)), Fₙ(x) is unbiased for F(x) with variance F(x)[1 − F(x)]/n. 1/(4n) is only the maximum of this variance, reached at the median.

Question 2 · easy

Counts are grouped into 7 classes and tested for fit to a Poisson distribution whose mean is estimated from the data. After pooling, 5 classes remain. How many degrees of freedom does Pearson's χ² have?

  1. 6
  2. 4
  3. 3
  4. 5
Answer and working

C. 3

df = (classes after pooling) − 1 − (parameters estimated) = 5 − 1 − 1 = 3. 5 counts classes before deductions; 6 uses the classes before pooling with one deduction.

Question 3 · easy

For n = 20 non-zero paired differences, what are the null mean and variance of the Wilcoxon signed-rank statistic T⁺?

  1. Mean 105, variance 717.5
  2. Mean 105, variance 2870
  3. Mean 10, variance 5
  4. Mean 210, variance 717.5
Answer and working

A. Mean 105, variance 717.5

E(T⁺) = n(n + 1)/4 = 105; Var(T⁺) = n(n + 1)(2n + 1)/24 = 717.5. 210 is the total of all ranks; mean 10 and variance 5 belong to the sign statistic.

Question 4 · medium

What does U₁/(n₁n₂) estimate in the Mann-Whitney setting, ignoring ties?

  1. The ratio of the two population variances
  2. The correlation between the two samples
  3. The difference between the two population medians (θ₁ − θ₂)
  4. P(X > Y) for X from sample 1 and Y from sample 2
Answer and working

D. P(X > Y) for X from sample 1 and Y from sample 2

U₁ counts pairs with X > Y, so U₁/(n₁n₂) is the unbiased estimator of P(X > Y), a scale-free effect size. It equals a median difference only under further assumptions.

Question 5 · medium

Three fungicides are applied to 5 plots each; the rank sums of percentage infection are 20, 40 and 60. What is the Kruskal-Wallis H?

  1. 8.0
  2. 232
  3. 56
  4. 4.0
Answer and working

A. 8.0

H = [12/(15 × 16)] × (400 + 1600 + 3600)/5 − 3 × 16 = 0.05 × 1120 − 48 = 8.0, with 2 df; it exceeds 5.991. Forgetting to divide by nᵢ gives 232.

Question 6 · hard

For k = 2 groups, how is the Kruskal-Wallis H related to the Mann-Whitney test?

  1. They are unrelated statistics
  2. H is half the Mann-Whitney U
  3. H equals the Mann-Whitney U itself, unstandardised
  4. H equals the square of the Mann-Whitney Z
Answer and working

D. H equals the square of the Mann-Whitney Z

With two groups H is the squared standardised rank-sum statistic, Z², and χ²₁ is the distribution of Z² for normal Z.

Question 7 · medium

In a randomised block trial the treatments are increasing doses, and a monotone dose effect is expected. Which rank test is designed for this?

  1. The Wald-Wolfowitz runs test
  2. Dunnett's test
  3. The one-sample KS test
  4. Page's test
Answer and working

D. Page's test

Page's L test uses within-block ranks and weights treatments by their hypothesised order; it is the blocked counterpart of Jonckheere-Terpstra.

Question 8 · easy

Yield and fertiliser rate are related by y = log(x) exactly for 8 plots with distinct x. What is Spearman's ρ_s?

  1. 0, because log is non-linear
  2. 1
  3. Less than 1, because the relation is not linear
  4. Equal to Pearson's r for these data
Answer and working

B. 1

The ranks of y and x agree perfectly for any strictly increasing relation, so ρ_s = 1. Pearson's r would be below 1 because the relation is not linear.

Question 9 · medium

Disease severity is scored weekly on each plot through the season. Which summary gives a quantitative plot-level response suited to ANOVA?

  1. The area under the disease progress curve
  2. The number of assessment dates
  3. The modal score over the season
  4. The score on the first assessment date only
Answer and working

A. The area under the disease progress curve

AUDPC integrates severity over time (trapezoidal rule) and behaves much more like a continuous variable than individual scores.

Question 10 · medium

Spearman's correlation between farmer preference rank and yield rank over 26 varieties is 0.45. Using Z = ρ_s√(n − 1), what is the large-sample test statistic for independence?

  1. 2.29
  2. 2.47
  3. 2.25
  4. 0.090
Answer and working

C. 2.25

Var(ρ_s) = 1/(n − 1) under H₀, so Z = 0.45 × √25 = 2.25, significant at 5%. The t form 0.45√24/√(1 − 0.45²) = 2.47 is an alternative approximation, not the Z asked for.

A study plan for Nonparametric & Rank-Based Methods in Indian Statistical Institute admission tests

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 56 questions): Mann-Whitney U test and two-sample Kolmogorov-Smirnov test (9, 16%); Asymptotic relative efficiency (7, 13%); Empirical distribution function and its properties (6, 11%); Friedman's test for randomised blocks (6, 11%).

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

  2. Then the middle band: Goodness of fit: chi-square and Kolmogorov-Smirnov tests (6, 11%); Sign test and Wilcoxon signed-rank test (5, 9%); Spearman's and Kendall's rank correlations (5, 9%); Kruskal-Wallis test (4, 7%).

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

  3. Then the lighter units: Analysing ordinal scores: disease ratings and sensory panels (3, 5%); Run test for randomness (3, 5%); Cochran's Q test (2, 4%).

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

  4. Leave time for the hard ones: Cochran's Q test (1 hard of 2), Empirical distribution function and its properties (2 hard of 6), Analysing ordinal scores: disease ratings and sensory panels (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: Empirical distribution function and its properties (4 easy of 6); Sign test and Wilcoxon signed-rank test (3 easy of 5).

    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 Institute admission tests 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.