ICAR-IASRI MSc and PhD Statistical Computing: Simulation, Resampling & Algorithms: practice questions
ICAR-IASRI MSc and PhD (Agricultural Statistics, Computer Application, Bioinformatics) · India · ICAR-Indian Agricultural Statistics Research Institute with the IARI Graduate School
Original practice questions written to the published syllabus · not past papers
10worked here, free
21written to this exam in this subject
11syllabus units
About ICAR-IASRI MSc and PhD
ICAR-IASRI MSc and PhD (Agricultural Statistics, Computer Application, Bioinformatics) is a postgraduate admission exam run by ICAR-Indian Agricultural Statistics Research Institute with the IARI Graduate School, for candidates in India. Admission through ICAR AIEEA-PG (Master's) and AICE-JRF/SRF (PhD) plus the Graduate School's own process.
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.
Statistical Computing: Simulation, Resampling & Algorithms in ICAR-IASRI MSc and PhD
The algorithms inside modern statistics: random-number generation, Monte Carlo integration, the bootstrap and jackknife, permutation tests, the EM algorithm, MCMC and numerical linear algebra, with attention to accuracy and reproducibility.
Our 21 ICAR-IASRI MSc and PhD questions in this subject: 11 easy, 10 medium, 0 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-IASRI MSc and PhD and are not shown on any other exam's page.
Question 1 · medium
A randomness test that a generator passes shows that:
No non-randomness was detected
The sequence is proven random for every purpose
The generator is cryptographically secure
The period is infinite
Answer and working
A. No non-randomness was detected
Tests can only fail to detect departures from randomness.
Question 2 · medium
Compared with a bootstrap standard error, a bootstrap confidence interval generally needs:
Fewer resamples
More resamples
No resamples at all
Exactly 50 resamples
Answer and working
B. More resamples
Tail percentiles need 1,000–2,000 or more resamples.
Question 3 · medium
PERMANOVA and the Mantel test are examples of:
Parametric ANOVA with normal errors
Bootstrap intervals
Permutation methods
MCMC samplers
Answer and working
C. Permutation methods
Both obtain null distributions by permutation.
Question 4 · medium
For multivariate normal data with missing values, EM differs from simply filling in predicted values because it:
Ignores observed values when imputing
Deletes incomplete rows
Adds conditional variances
Uses medians
Answer and working
C. Adds conditional variances
Single imputation of predictions understates variability.
Question 5 · medium
The initial iterations discarded from an MCMC run are called:
Thinning
Burn-in or warm-up
Proposals
Effective samples
Answer and working
B. Burn-in or warm-up
They depend on starting values.
Question 6 · easy
Statistical software solves least squares preferably by:
Explicitly inverting X′X
Cramer's rule
QR decomposition of X
Random search
Answer and working
C. QR decomposition of X
QR avoids squaring the condition number.
Question 7 · medium
The standard, most stable way to compute a dense QR decomposition is with:
Householder reflections
Cramer's rule
Classical Gram–Schmidt without reorthogonalisation
Repeated squaring
Answer and working
A. Householder reflections
Givens rotations suit sparse matrices and updating.
Question 8 · easy
In the ADEMP framework for simulation studies, the 'E' stands for:
Data-generating mechanisms in a good simulation study should include:
Only the scenario favouring the new method
A single sample size
No departures from assumptions
Realistic and challenging scenarios
Answer and working
D. Realistic and challenging scenarios
Factorial scenarios probe where methods break.
Question 10 · medium
GPUs accelerate statistical computing mainly for:
Reading text files from disk quickly
Sequential loops with branching
Printing results
The same arithmetic on many data
Answer and working
D. The same arithmetic on many data
Thousands of simple cores suit array operations.
A study plan for Statistical Computing: Simulation, Resampling & Algorithms in ICAR-IASRI MSc and PhD
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.
First, the heavy units (38% of the 21 questions): Bootstrap: percentile and BCa intervals (2, 10%); EM algorithm and its applications (missing data, mixtures) (2, 10%); Generating random variates: inversion, acceptance-rejection and transformation methods (2, 10%); Jackknife (2, 10%).
Read the manual chapters for these first and do every worked question you can find on them.
Then the middle band: MCMC algorithms in practice (2, 10%); Monte Carlo integration and variance reduction (2, 10%); Numerical linear algebra for statistics: QR and Cholesky (2, 10%); Permutation and randomisation tests (2, 10%).
Aim for one sitting per unit, then mixed questions across them.
Then the lighter units: Pseudo-random number generation and tests of randomness (2, 10%); Simulation studies: design and reporting (2, 10%); Parallel and high-performance computing for statistics (1, 5%).
Learn the definitions, classifications and standard numbers; these units are more often recalled than worked.
Bank the easy marks: Simulation studies: design and reporting (2 easy of 2); Bootstrap: percentile and BCa intervals (1 easy of 2).
Do these early in revision so they are fast on the day.
Finish timed. Sit the subject as a timed set without notes, mark it, and send each wrong answer back to its unit above.