LAISHA
Open Laisha

IIT JAM Multivariate Analysis: practice questions

IIT JAM - Mathematical Statistics (MS) · India · IITs (JAM 2026 organised by IIT Bombay)

Original practice questions written to the published syllabus · not past papers
10worked here, free
73written to this exam in this subject
12syllabus units

About IIT JAM

IIT JAM - Mathematical Statistics (MS) is a postgraduate admission exam run by IITs (JAM 2026 organised by IIT Bombay), for candidates in India. Computer-based, 3 hours, 60 questions, 100 marks, English. Section A 30 MCQs (10 x 1 mark, 20 x 2 marks); Section B 10 MSQs (2 marks, no negative or partial marking); Section C 20 NAT (10 x 1, 10 x 2, no negative). MCQ wrong answers lose 1/3 or 2/3 mark.

Source: jam2026.iitb.ac.in.

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

Multivariate Analysis in IIT JAM

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 73 IIT JAM questions in this subject: 20 easy, 44 medium, 9 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 IIT JAM and are not shown on any other exam's page.

Question 1 · medium

Two traits keep the same variances, but their correlation rises from 0.5 to 0.9. What happens to the generalised variance |S|?

  1. It increases
  2. It becomes negative
  3. It stays the same
  4. It decreases
Answer and working

D. It decreases

|S| = s₁²s₂²(1 − r²), which shrinks as |r| grows: the scatter ellipse flattens. The total variance tr(S) would be unchanged. |S| cannot be negative for a valid dispersion matrix.

Question 2 · easy

What is E(W) for W ~ Wₚ(m, Σ)?

  1. Σ
  2. Σ/m
  3. mΣ
  4. (m − p − 1)Σ
Answer and working

C. mΣ

W = ΣZᵢZᵢ′ with E(ZᵢZᵢ′) = Σ, so E(W) = mΣ, as E(σ²χ²_m) = mσ² for p = 1. (m − p − 1)Σ appears in E(W⁻¹) = Σ⁻¹/(m − p − 1).

Question 3 · medium

Which distribution is the conjugate prior for Σ in Bayesian analysis of multivariate normal data?

  1. Multivariate normal
  2. Wishart on Σ itself
  3. Inverse Wishart
  4. Dirichlet
Answer and working

C. Inverse Wishart

The inverse-Wishart prior on Σ (equivalently Wishart on Σ⁻¹) gives an inverse-Wishart posterior, the matrix version of the inverse-gamma prior for a variance.

Question 4 · medium

The non-zero eigenvalues of W⁻¹B are 1.0 and 0.25. What is Pillai's trace?

  1. 1.00
  2. 0.70
  3. 1.25
  4. 0.40
Answer and working

B. 0.70

V = Σλᵢ/(1 + λᵢ) = 0.5 + 0.2 = 0.70. 1.25 is the Hotelling-Lawley trace; 0.40 is Wilks' Λ; 1.00 is Roy's largest root.

Question 5 · easy

What is the variance of the j-th principal component of Σ?

  1. σⱼⱼ, the j-th diagonal element
  2. λⱼ, the j-th eigenvalue of Σ
  3. λⱼ², the squared eigenvalue
  4. 1, after standardisation
Answer and working

B. λⱼ, the j-th eigenvalue of Σ

Var(aⱼ′X) = aⱼ′Σaⱼ = λⱼ when aⱼ is the unit eigenvector for λⱼ.

Question 6 · hard

A covariance matrix has variances 9 and 4 and covariance 2. What proportion of total variance does the first principal component explain?

  1. 50.0%
  2. 107.8%
  3. 69.2%
  4. 74.6%
Answer and working

D. 74.6%

Eigenvalues solve λ² − 13λ + 32 = 0: λ = 9.702 and 3.298. Proportion = 9.702/13 = 74.6%. 69.2% is the share of the first variable alone, ignoring the covariance.

Question 7 · easy

What does Ward's method minimise at each merge?

  1. The smallest distance between any two members
  2. The number of clusters
  3. Increase in within-cluster SS
  4. The largest distance between any two members
Answer and working

C. Increase in within-cluster SS

Ward merges the pair of clusters whose union increases the error sum of squares least, giving compact, similar-sized clusters.

Question 8 · medium

What does the cophenetic correlation measure?

  1. Correlation between variables within a cluster
  2. Correlation between the two largest clusters
  3. Dendrogram fidelity to original distances
  4. Agreement between two different k-means runs
Answer and working

C. Dendrogram fidelity to original distances

Values close to 1 mean the tree represents the original dissimilarities faithfully; low values warn that the dendrogram distorts the data.

Question 9 · hard

For uncorrelated traits with standard deviations σⱼ, what does D² between two population means reduce to?

  1. (Σδⱼ/σⱼ)²
  2. Σδⱼ/σⱼ²
  3. Σ(δⱼ/σⱼ)²
  4. Σδⱼ²
Answer and working

C. Σ(δⱼ/σⱼ)²

With diagonal Σ, Σ⁻¹ has entries 1/σⱼ², so D² = Σδⱼ²/σⱼ²: the sum of squared standardised differences.

Question 10 · easy

Classical multidimensional scaling of Euclidean distances computed from a data matrix gives the same configuration as which method?

  1. Principal component analysis
  2. Factor analysis with varimax
  3. Canonical correlation
  4. Fisher's discriminant analysis
Answer and working

A. Principal component analysis

Principal coordinate analysis of Euclidean distances reproduces the PCA scores (up to sign and rotation).

A study plan for Multivariate Analysis in IIT JAM

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 (38% of the 73 questions): Cluster analysis: hierarchical and k-means (7, 10%); Discriminant analysis and classification (7, 10%); Estimation of the mean vector and dispersion matrix (7, 10%); Multivariate normal distribution: properties, marginals and conditionals (7, 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 (6, 8%); Factor analysis (6, 8%); Hotelling's T-squared and its applications (6, 8%); Multidimensional scaling and correspondence analysis (6, 8%).

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

  3. Then the lighter units: Multivariate analysis of variance (MANOVA) (6, 8%); Principal component analysis (6, 8%); Wishart distribution (5, 7%); Mahalanobis D-squared and genetic divergence (4, 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) (2 hard of 6), Discriminant analysis and classification (2 hard of 7), Mahalanobis D-squared and genetic divergence (1 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: Estimation of the mean vector and dispersion matrix (3 easy of 7); Wishart distribution (2 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 — IIT JAM 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.