LAISHA
Open Laisha

ICAR AICE-JRF/SRF - Agricultural Statistics, Computer Application, Bioinformatics (PhD admission and fellowship)

India · National Testing Agency for ICAR

2601practice questions written to this syllabus
48subjects

What ICAR AICE-JRF/SRF is

ICAR AICE-JRF/SRF - Agricultural Statistics, Computer Application, Bioinformatics (PhD admission and fellowship) is a PhD, fellowship and lectureship exam run by National Testing Agency for ICAR, for candidates in India.

Format

Computer-based test, 2 hours, 120 MCQs, +4 / -1 (Part A general, Part B core group, Part C subject). 2026 exam held with AIEEA-PG in July 2026.

Sections

How it has changed

Unit lists read in the ICAR AICE-JRF/SRF syllabus PDF (https://icar.org.in/sites/default/files/Circulars/Syllabus%20AICE%20JRF_SRF%20(PhD).pdf). The core-group number that holds these subjects in the 2026 bulletin could not be read (the bulletin PDF is a scanned image), so the Part B core-group content is not verified.

How to prepare for ICAR AICE-JRF/SRF

Where our 2601 questions for it fall

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.

  1. Heaviest (54% of the questions): Statistical Genomics: QTL Mapping, GWAS & Genomic Selection (100); Crop Forecasting, Weather-Based Models & Non-linear Growth Models (98); Principles of Experimental Design & Basic Designs (96); Bioinformatics & Computational Biology (95); Linear Models & Regression Analysis (92); Sampling Theory II: Unequal Probability, Cluster, Multistage & Repeated Sampling (89); Factorial Experiments, Confounding, Split- & Strip-Plot Designs (88); Statistical Genetics: Population & Quantitative Genetics (87); Groups of Experiments, Multi-Location Trials & Genotype x Environment Analysis (86); Advanced Design of Experiments (PhD) (86); Incomplete Block Designs & Design Construction (84); Biometrical Techniques in Plant & Animal Breeding (84); and 4 more.

    Start here: read these subjects' manuals first and work their questions until the reasoning is automatic.

  2. Middle band: Survey Design, Questionnaires, Field Operations & Non-Sampling Errors (73); Crop Area & Yield Estimation: Crop-Cutting Experiments (70); Bayesian Inference (67); Categorical Data Analysis & Generalized Linear Models (65); Optimization Techniques for Statisticians (64); Nonparametric & Rank-Based Methods (60); Sampling Theory I: Simple, Stratified, Systematic, Ratio & Regression Estimation (57); Bioassay, Survival & Epidemiological Statistics (54); Statistical Inference I: Estimation (51); Time Series Analysis (49); Computer Science Fundamentals: Architecture, Data Structures, Networks & Operating Systems (48); Probability Distributions & Sampling Distributions (48); and 4 more.

    One pass per subject, then mixed sets across them.

  3. Lighter: Stochastic Processes (40); Databases, Big Data & Agricultural Information Systems (34); Agricultural Statistics System, Census & Official Statistics (33); Statistical Methods for Agricultural Sciences (28); Index Numbers, National Accounts & Agricultural Economic Statistics (26); Exploratory Data Analysis & Statistical Graphics (25); Data Analysis with Statistical Software & Reproducible Research (24); Statistical Consulting, Research Methodology & Scientific Communication (19); Statistical Quality Control & Reliability (17); Programming for Statistics & Data (R and Python) (16); Differential Equations, Finite Differences & Numerical Analysis (15); Linear Algebra & Matrix Theory for Statistics (12); and 4 more.

    Definitions, classifications and standard figures; quick to revise, easy marks to lose.

  4. Hardest share: Statistical Quality Control & Reliability (13 hard of 17), Linear Algebra & Matrix Theory for Statistics (9 hard of 12), Statistical Methods for Agricultural Sciences (19 hard of 28).

    Leave untimed time for these and read every explanation, including for the options you rejected.

Official source: icar.org.in

8 practice questions, worked

Original questions written to this exam's published syllabus and pattern. Not past papers.

1. Following Gomez and Gomez, germination percentages that all lie between 35% and 65% should be:

  1. Given the arcsine transformation
  2. Given the logarithmic transformation
  3. Analysed without transformation
  4. Given the square-root transformation
Answer and working

C. Analysed without transformation

Binomial variance changes little in the 30-70% range, so no transformation is needed. Square root is suggested when data lie wholly within 0-30% or 70-100%, and arcsine when they span a wider range.

Principles of Experimental Design & Basic Designs · Assumptions of ANOVA and variance-stabilising transformations

2. A one-way trial has 3 varieties with true effects −2, 0 and 2 kg, 4 replicates each and error variance 4 kg². With λ = rΣτᵢ²/σ², what is the non-centrality parameter of the treatment F-statistic?

  1. 16
  2. 2
  3. 8
  4. 4
Answer and working

C. 8

λ = 4 × (4 + 0 + 4)/4 = 8. Omitting r gives 2; dividing by σ = 2 instead of σ² gives 16; using |τ| instead of τ² gives 4.

Probability Distributions & Sampling Distributions · Non-central chi-square, t and F

3. A random forest forecasts district yields with very high accuracy under random 10-fold cross-validation, but performs poorly in the next season. What is the most likely reason?

  1. The next season had no weather data at all
  2. Random forests cannot use weather variables
  3. Cross-validation always underestimates accuracy
  4. Folds leaked correlated years and districts
Answer and working

D. Folds leaked correlated years and districts

Yields are correlated in time and space; random folds put the same year or nearby districts in both training and test sets, inflating accuracy. Leave-one-year-out or forward validation is honest.

Crop Forecasting, Weather-Based Models & Non-linear Growth Models · Machine-learning and hybrid forecasting models

4. What is the adjusted total Q_i of treatment i in a block design?

  1. T_i plus the sum of the totals of blocks not containing treatment i
  2. T_i minus the sum of (block total / block size) over its blocks
  3. T_i minus (grand mean x r_i), ignoring which blocks held treatment i
  4. T_i divided by the number of blocks in which treatment i appears
Answer and working

B. T_i minus the sum of (block total / block size) over its blocks

Q = T - N K^(-1) B: each block containing the treatment contributes its total divided by its size.

Incomplete Block Designs & Design Construction · General block design: C-matrix, connectedness and balance

5. What is the resolution of a 2^(4-1) design with defining relation I = ABCD?

  1. V
  2. III
  3. IV
  4. II
Answer and working

C. IV

Resolution is the length of the shortest word in the defining relation; ABCD has four letters, so resolution IV: main effects are clear of two-factor interactions, but two-factor interactions are aliased with each other.

Factorial Experiments, Confounding, Split- & Strip-Plot Designs · Fractional factorials and aliasing

6. Under which log-linear model can the XY association be studied in the XY marginal table without bias?

  1. (XY, XZ) — Z is conditionally independent of Y given X
  2. (XY, XZ, YZ) — homogeneous association in all pairs
  3. (XYZ) — the saturated model
  4. (XZ, YZ) — X and Y are conditionally independent given Z
Answer and working

A. (XY, XZ) — Z is conditionally independent of Y given X

Collapsibility over Z holds when Z is conditionally independent of X or of Y; under (XY, XZ) the conditional and marginal XY odds ratios coincide. Under (XZ, YZ) the marginal table can show association that is absent conditionally.

Categorical Data Analysis & Generalized Linear Models · Log-linear models for multi-way tables

7. Cost C2 equals Cost B2 plus:

  1. Rent on leased-in land
  2. 10% managerial cost
  3. Imputed value of family labour
  4. Land revenue
Answer and working

C. Imputed value of family labour

C2 = B2 + imputed family labour.

Agricultural Statistics System, Census & Official Statistics · Comprehensive scheme for cost of cultivation

8. A mixed model reports non-convergence. Which cause is most common?

  1. A bug in every optimiser
  2. Over-parameterisation
  3. Too many observations
  4. Normal residuals
Answer and working

B. Over-parameterisation

Unidentifiable or boundary parameters make the objective flat or constrained; simplifying or reparameterising usually helps.

Optimization Techniques for Statisticians · Numerical optimisation: gradient, Newton and quasi-Newton methods

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 all 2,601 questions — ICAR AICE-JRF/SRF - Agricultural Statistics, Computer Application, Bioinformatics (PhD adm

US$10 a month · one price worldwide · cancel any time

  • Every practice question for this course, with explanations
  • Timed papers for the exams this course leads to
  • Ask Laisha to explain any topic — 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.