LAISHA
Open Laisha

Indian Statistical Institute admission tests Multivariate Analysis: 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
59written to this exam in this subject
12syllabus 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.

Multivariate Analysis in Indian Statistical Institute admission tests

Analysing many correlated variables together: the multivariate normal and its estimation, Wishart and Hotelling's T-squared, MANOVA, and the reduction and classification tools agricultural scientists use daily (principal components, factor analysis, discriminant and cluster analysis, Mahalanobis D-squared).

Our 59 Indian Statistical Institute admission tests questions in this subject: 20 easy, 37 medium, 2 hard, across 12 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

From n p-variate normal observations, what is the MLE of Σ?

  1. nS/(n − 1)
  2. (n − 1)S/n
  3. S
  4. S/n
Answer and working

B. (n − 1)S/n

The MLE divides the SSCP matrix by n, giving (n − 1)S/n; S (divisor n − 1) is the unbiased estimator. S/n estimates the dispersion of x̄.

Question 2 · easy

For p = 1, how is Hotelling's T² related to Student's t?

  1. T² = t²
  2. T² = √t
  3. T² = 2t
  4. T² = t
Answer and working

A. T² = t²

With one variable, S⁻¹ = 1/s², so T² = n(x̄ − μ₀)²/s² = t², and the F(1, n − 1) reference distribution is that of t².

Question 3 · easy

Canonical correlation analysis with one variable in the second set reduces to which quantity?

  1. Kendall's coefficient of concordance
  2. The simple correlation coefficient
  3. The multiple correlation coefficient
  4. The partial correlation coefficient
Answer and working

C. The multiple correlation coefficient

With q = 1, the best linear combination of the first set correlates with the single variable at the multiple correlation R; with p = q = 1 it is the simple correlation.

Question 4 · easy

For two groups with a common dispersion matrix, what is Fisher's linear discriminant function coefficient vector?

  1. (x̄₁ − x̄₂)/tr(S_p)
  2. S_p⁻¹(x̄₁ − x̄₂)
  3. S_p(x̄₁ − x̄₂)
  4. The first eigenvector of S_p
Answer and working

B. S_p⁻¹(x̄₁ − x̄₂)

Maximising between-group to within-group variance of a′x gives a ∝ S_p⁻¹(x̄₁ − x̄₂). The first eigenvector of S_p is the first principal component, which ignores the group difference.

Question 5 · hard

With priors p₁, p₂ and costs c(1 | 2), c(2 | 1), when does the minimum-expected-cost rule assign x to group 1?

  1. When p₁ > p₂, whatever the data
  2. When f₁(x) > f₂(x), whatever the priors
  3. When f₁(x)/f₂(x) > [c(2 | 1)p₁]/[c(1 | 2)p₂]
  4. When f₁(x)/f₂(x) > [c(1 | 2)p₂]/[c(2 | 1)p₁]
Answer and working

D. When f₁(x)/f₂(x) > [c(1 | 2)p₂]/[c(2 | 1)p₁]

Expected cost is minimised by comparing the density ratio with the ratio of prior-weighted costs of the two errors; with equal priors and costs the threshold is 1.

Question 6 · medium

Distances are d(A, B) = 2, d(A, C) = 6 and d(B, C) = 5. After A and B merge, what is the complete-linkage distance from cluster AB to C?

  1. 2
  2. 5.5
  3. 5
  4. 6
Answer and working

D. 6

Complete linkage uses the largest member distance, max(6, 5) = 6. Single linkage gives 5 and average linkage 5.5.

Question 7 · medium

Germplasm accessions are clustered on yield (kg/ha) and days to flowering using Euclidean distance on the raw data. What is the problem?

  1. Euclidean distance cannot be used for two variables
  2. Clustering requires multivariate normality
  3. Days to flowering must be transformed to logarithms
  4. Yield's large numerical scale dominates the distances; standardise first
Answer and working

D. Yield's large numerical scale dominates the distances; standardise first

Without standardisation, a variable measured in large units swamps the others. Clustering does not need normality.

Question 8 · medium

Two traits have unit variances and correlation 0.5. Two genotypes differ by 1 unit in each trait. What is the Mahalanobis D² between them?

  1. 1.00
  2. 2.00
  3. 1.41
  4. 1.33
Answer and working

D. 1.33

D² = (δ₁² − 2ρδ₁δ₂ + δ₂²)/(1 − ρ²) = (1 − 1 + 1)/0.75 = 1.33. Ignoring the correlation gives 2.00, counting the correlated difference twice.

Question 9 · medium

What does non-metric multidimensional scaling preserve, and what does it minimise?

  1. The exact distances; it minimises the trace
  2. The rank order of dissimilarities; it minimises stress
  3. The correlations; it maximises the eigenvalues
  4. The cluster memberships; it minimises within-cluster SS
Answer and working

B. The rank order of dissimilarities; it minimises stress

Kruskal's non-metric MDS fits a configuration whose distances are monotonically related to the dissimilarities, minimising stress.

Question 10 · medium

Which distance does correspondence analysis use to compare row profiles?

  1. Mahalanobis distance
  2. Euclidean distance on raw counts
  3. Jaccard distance
  4. Chi-square distance
Answer and working

D. Chi-square distance

Profiles are compared with each column weighted by the inverse of its overall proportion, the chi-square distance, which is why total inertia is χ²/n.

A study plan for Multivariate Analysis 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 (44% of the 59 questions): Cluster analysis: hierarchical and k-means (7, 12%); Estimation of the mean vector and dispersion matrix (7, 12%); Factor analysis (6, 10%); Multidimensional scaling and correspondence analysis (6, 10%).

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

  2. Then the middle band: Canonical correlation analysis (5, 8%); Discriminant analysis and classification (5, 8%); Wishart distribution (5, 8%); Multivariate analysis of variance (MANOVA) (4, 7%).

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

  3. Then the lighter units: Multivariate normal distribution: properties, marginals and conditionals (4, 7%); Principal component analysis (4, 7%); Hotelling's T-squared and its applications (3, 5%); Mahalanobis D-squared and genetic divergence (3, 5%).

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

  4. Leave time for the hard ones: Multivariate analysis of variance (MANOVA) (1 hard of 4), Discriminant analysis and classification (1 hard of 5).

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

  5. Bank the easy marks: Hotelling's T-squared and its applications (2 easy of 3); Multivariate normal distribution: properties, marginals and conditionals (2 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 — Indian Statistical Institute admission tests Multivariate Analysis

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.