LAISHA
Open Laisha

Indian Statistical Service Examination Nonparametric & Rank-Based Methods: 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
25written 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.

Nonparametric & Rank-Based Methods in Indian Statistical Service Examination

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 25 Indian Statistical Service Examination questions in this subject: 6 easy, 11 medium, 8 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 · medium

Which theorem states that sup over x of |Fₙ(x) − F(x)| → 0 almost surely, for any F?

  1. The Lindeberg-Lévy central limit theorem
  2. The Glivenko-Cantelli theorem
  3. The Neyman-Pearson lemma
  4. Khintchine's weak law of large numbers
Answer and working

B. The Glivenko-Cantelli theorem

Glivenko-Cantelli gives uniform almost-sure convergence of the EDF. The strong law gives convergence at each fixed x; uniformity is the extra content.

Question 2 · medium

Which statement compares Pearson's chi-square and the Kolmogorov-Smirnov goodness-of-fit tests correctly?

  1. KS ignores the order of the classes; chi-square uses that order in its statistic
  2. KS uses ungrouped continuous data; chi-square suits counts and grouped data
  3. Chi-square needs a fully specified F₀; KS allows estimated parameters
  4. Chi-square is exact in small samples; KS is only an approximation
Answer and working

B. KS uses ungrouped continuous data; chi-square suits counts and grouped data

KS is exact and distribution-free for continuous, fully specified F₀ and avoids grouping. Chi-square handles discrete data and estimated parameters (with df adjustment) but loses information by grouping and ignores the order of classes.

Question 3 · medium

What is a weakness of the Wald-Wolfowitz two-sample runs test compared with the Mann-Whitney test?

  1. It cannot detect differences in spread
  2. It is valid only for equal sample sizes
  3. It needs normally distributed data
  4. Low power against a pure location shift
Answer and working

D. Low power against a pure location shift

The runs test reacts to any difference in distribution, but for a location shift Mann-Whitney is much more powerful. Neither needs normality, and the runs test does detect spread differences.

Question 4 · easy

What is the two-sample Kolmogorov-Smirnov statistic?

  1. The number of runs in the pooled sample
  2. sup over x of |F₁(x) − F₂(x)|
  3. The sum of ranks of the first sample (W₁)
  4. The difference in sample medians
Answer and working

B. sup over x of |F₁(x) − F₂(x)|

D = sup|F₁ − F₂|, the largest gap between the two EDFs. It detects differences of any kind, not only location.

Question 5 · hard

With heavy ties among disease scores, how is the Kruskal-Wallis H corrected?

  1. H is multiplied by 1 − Σ(t³ − t)/(N³ − N)
  2. Each tied value is dropped before ranking
  3. The df are reduced by the number of tie groups
  4. H is divided by 1 − Σ(t³ − t)/(N³ − N), enlarging it
Answer and working

D. H is divided by 1 − Σ(t³ − t)/(N³ − N), enlarging it

Mid-ranks reduce the variance of the ranks; dividing by 1 − Σ(t³ − t)/(N³ − N) restores the correct scale, so the corrected H is larger. Dropping ties discards data.

Question 6 · medium

Under normal errors, what is the asymptotic relative efficiency of the Kruskal-Wallis test relative to the one-way ANOVA F-test?

  1. 0.912
  2. 0.955
  3. 1.000
  4. 0.637
Answer and working

B. 0.955

3/π ≈ 0.955, the same as the Mann-Whitney/t ARE. 2/π is the sign test's ARE; 9/π² belongs to rank correlation tests.

Question 7 · easy

For k = 2 treatments in b blocks, Friedman's test reduces to which test?

  1. The sign test
  2. McNemar's test
  3. The Kruskal-Wallis test
  4. The Mann-Whitney test
Answer and working

A. The sign test

Ranking two treatments within each block records only which one is larger: exactly the information used by the sign test on paired differences.

Question 8 · hard

Under bivariate normality, what is the asymptotic relative efficiency of tests of independence based on Spearman's or Kendall's coefficient, relative to the test based on Pearson's r?

  1. 1.000
  2. 0.955
  3. 0.637
  4. 0.912
Answer and working

D. 0.912

9/π² ≈ 0.912: about 9% efficiency lost under normality, in exchange for robustness and invariance to monotone transformations.

Question 9 · hard

What is the asymptotic relative efficiency of the Wilcoxon signed-rank test relative to the t-test for uniform data?

  1. 0.333
  2. 0.955
  3. 1
  4. 1.5
Answer and working

C. 1

For the uniform distribution the ARE is 1. 1/3 is the sign/t ARE for the uniform; 1.5 is the Wilcoxon/t ARE for the Laplace.

Question 10 · hard

For a continuous F, what is the limiting distribution of √n Dₙ, where Dₙ = sup|Fₙ − F|?

  1. Chi-square with 1 degree of freedom, as for a single cell
  2. The standard normal distribution (by the central limit theorem)
  3. Kolmogorov's distribution, 1 − 2Σ(−1)^(k−1) e^(−2k²x²)
  4. The exponential distribution with mean 1/2, as for a maximum
Answer and working

C. Kolmogorov's distribution, 1 − 2Σ(−1)^(k−1) e^(−2k²x²)

√n(Fₙ − F) converges to a Brownian bridge, and the supremum of its absolute value has Kolmogorov's distribution, P(K ≤ x) = 1 − 2Σ_(k≥1)(−1)^(k−1)exp(−2k²x²). Its 95% point is about 1.36, the source of the 1.36/√n critical value.

A study plan for Nonparametric & Rank-Based Methods 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 25 questions): Asymptotic relative efficiency (5, 20%); Kruskal-Wallis test (3, 12%); Run test for randomness (3, 12%); Cochran's Q test (2, 8%).

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

  2. Then the middle band: Empirical distribution function and its properties (2, 8%); Friedman's test for randomised blocks (2, 8%); Goodness of fit: chi-square and Kolmogorov-Smirnov tests (2, 8%); Mann-Whitney U test and two-sample Kolmogorov-Smirnov test (2, 8%).

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

  3. Then the lighter units: Spearman's and Kendall's rank correlations (2, 8%); Analysing ordinal scores: disease ratings and sensory panels (1, 4%); Sign test and Wilcoxon signed-rank test (1, 4%).

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

  4. Leave time for the hard ones: Kruskal-Wallis test (2 hard of 3), Empirical distribution function and its properties (1 hard of 2), Friedman's test for randomised blocks (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: Cochran's Q test (1 easy of 2); Friedman's test for randomised blocks (1 easy of 2).

    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 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.