India · National Testing Agency for CSIR · unverified source
CSIR-UGC NET - Mathematical Sciences (statistics option) is a PhD, fellowship and lectureship exam run by National Testing Agency for CSIR, for candidates in India.
Computer-based, 3 hours, Parts A (general aptitude), B and C (subject); statisticians answer from the statistics unit.
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: csirnet.nta.ac.in.
Qualifies for JRF and lectureship in mathematical sciences, including statistics departments of universities. The NTA CSIR-NET page fetched did not show the syllabus.
Our reading of the syllabus, from the questions we have written to this exam; the exam body's own weighting, where published, is under Format above.
Start here: read these subjects' manuals first and work their questions until the reasoning is automatic.
One pass per subject, then mixed sets across them.
Definitions, classifications and standard figures; quick to revise, easy marks to lose.
Leave untimed time for these and read every explanation, including for the options you rejected.
Official source: csirnet.nta.ac.in
Original questions written to this exam's published syllabus and pattern. Not past papers.
1. In what situation can DIC be unreliable, even giving a negative effective number of parameters p_D?
A. When the posterior is strongly non-normal or multimodal, as in mixtures
p_D = D̄ − D(θ̄) depends on the deviance at the posterior mean, which can be a poor point for skewed or multimodal posteriors; WAIC and PSIS-LOO avoid this.
Bayesian Inference · Posterior predictive checks and model comparison (DIC, WAIC)
2. A factor analysis reports an estimated uniqueness of −0.05 for one variable. What is this called and what does it suggest?
C. A Heywood case
Variances cannot be negative; a zero or negative uniqueness (Heywood case) indicates an improper solution, from over-factoring, small samples or near-collinear variables.
Multivariate Analysis · Factor analysis
3. Counts are Poisson(λ) with a gamma prior on λ. What is the posterior predictive distribution of a new count?
B. Negative binomial
Mixing a Poisson over a gamma distribution for its mean gives a negative binomial; the extra variance reflects uncertainty about λ. Beta-binomial is the predictive for binomial data with a beta prior.
Bayesian Inference · Conjugate families
4. A slope estimate is 2.0 with standard error 0.4 and t = 2.0. What is Fieller's g?
C. 0.16
g = t²v₂₂/b² = 4 × 0.16/4 = 0.16, large enough for Fieller limits to differ noticeably from delta-method limits. 0.40 is t × SE/b; 0.64 omits division by b².
Bioassay, Survival & Epidemiological Statistics · Parallel-line and slope-ratio assays; Fieller's theorem
5. A periodic irreducible positive-recurrent chain has stationary distribution π. Which statement is correct?
C. Long-run time fractions equal π, though pᵢⱼ⁽ⁿ⁾ itself does not converge
Periodicity prevents convergence of n-step probabilities, but averages over time still converge to π (the ergodic theorem needs no aperiodicity).
Stochastic Processes · Limiting and stationary distributions
6. For a Poisson GLM with the canonical log link, which statistic is sufficient for β?
D. X′y
With θ = η = Xβ the log-likelihood is β′X′y − Σexp(xᵢ′β) − Σlog yᵢ!, so X′y is sufficient.
Categorical Data Analysis & Generalized Linear Models · Poisson regression for counts
7. In measure-theoretic terms, the probability density function of a continuous random variable X is:
D. The Radon-Nikodym derivative of the law P_X with respect to Lebesgue measure
P_X is absolutely continuous with respect to Lebesgue measure λ, so the Radon-Nikodym theorem gives f = dP_X/dλ with P(X ∈ B) = ∫_B f dλ. The reverse derivative dλ/dP_X need not even exist. A pmf is the corresponding derivative with respect to counting measure.
Probability Theory · Elements of measure theory: measurable functions and Lebesgue integral
8. Why are non-central t and F distributions needed in planning field experiments?
A. They give the distribution of the test statistic under the alternative, from which power and replication follow
Under H₀ the statistic is central; under an alternative it is non-central with non-centrality set by the true effects, σ² and replication. Power = P(non-central variable > central critical value). Estimated error variance is already built into the central t and F.
Probability Distributions & Sampling Distributions · Non-central chi-square, t and F
Last updated · How Laisha writes and checks its pages · Report a mistake
US$10 a month · one price worldwide · cancel any time
| Free | Smarter | |
|---|---|---|
| Grow guides, pest pages, articles | Read all | Read all |
| Ask Laisha | 5 a day | Up to 1,000 a month |
| Exam practice | The sample on each page | Every question, timed papers |
| Crop calendar and pest alerts | — | For your own fields |
| Calculators and tools | — | All |
| Uploads and reports | — | 100 a month, real files |
An AI operating system for agriculture: free guides, manuals and tools written for every country.