LAISHA
Open Laisha

ICAR AIEEA-PG Statistical Inference II: Testing of Hypotheses & Sequential Analysis: practice questions

ICAR AIEEA-PG - Group 07 Statistical Sciences (Master's admission and PG scholarship) · India · National Testing Agency for ICAR

Original practice questions written to the published syllabus · not past papers
10worked here, free
35written to this exam in this subject
8syllabus units

About ICAR AIEEA-PG

ICAR AIEEA-PG - Group 07 Statistical Sciences (Master's admission and PG scholarship) is a postgraduate admission exam run by National Testing Agency for ICAR, for candidates in India. Computer-based test, 120 minutes, single paper of 120 MCQs, +4 / -1, English. 2026 exam on 4 July 2026. Fills ICAR's all-India quota of Master's seats at agricultural universities.

Source: cdnbbsr.s3waas.gov.in.

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

Statistical Inference II: Testing of Hypotheses & Sequential Analysis in ICAR AIEEA-PG

Two types of error, level of significance, and the z, t, chi-square and F tests applied to agricultural data.

The logic and optimality of tests: errors and power, the Neyman-Pearson lemma, uniformly most powerful and unbiased tests, likelihood-ratio tests and their large-sample forms, and Wald's sequential probability ratio test.

Our 35 ICAR AIEEA-PG questions in this subject: 22 easy, 12 medium, 1 hard, across 8 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 AIEEA-PG and are not shown on any other exam's page.

Question 1 · easy

In testing whether a new fertiliser raises yield, what is a Type I error?

  1. Choosing a significance level that is larger than 0.05
  2. Concluding it raises yield when it does not
  3. Using a one-sided test when a two-sided test was needed
  4. Concluding that it does not raise yield when in fact it does
Answer and working

B. Concluding it raises yield when it does not

A Type I error rejects a true null hypothesis (here, 'no increase'). Missing a real increase is a Type II error. The significance level is the probability of a Type I error, not the error itself.

Question 2 · easy

Plot yields are N(μ, 16). From 16 plots, H₀: μ = 50 is rejected when x̄ > 51. What is the size of the test?

  1. 0.0500
  2. 0.1587
  3. 0.3173
  4. 0.4013
Answer and working

B. 0.1587

Under H₀, X̄ ~ N(50, 1), so α = P(X̄ > 51) = P(Z > 1) = 0.1587. Using σ = 4 instead of SE = 1 gives z = 0.25; doubling gives a two-sided figure that does not apply to this one-sided rule.

Question 3 · medium

A variety trial reports p = 0.03 for the difference between two varieties. Which statement is correct?

  1. The probability that there is no true difference between the two varieties is 0.03
  2. There is a 97% probability that the new variety truly out-yields the old one
  3. If there were no true difference, a result at least this extreme would occur with probability 0.03
  4. The probability that the observed difference is due to random chance alone is exactly 0.03 here
Answer and working

C. If there were no true difference, a result at least this extreme would occur with probability 0.03

The p-value is computed assuming H₀ is true. It is not the probability of H₀, nor of 'chance', and 1 − p is not the probability of H₁; those need a Bayesian analysis with a prior.

Question 4 · easy

A two-sided z-test gives z = 2.2. What is the p-value?

  1. 0.9722
  2. 0.0278
  3. 0.0139
  4. 0.9861
Answer and working

B. 0.0278

p = 2P(Z > 2.2) = 2 × 0.0139 = 0.0278. 0.0139 is the one-sided p-value.

Question 5 · easy

X₁, ..., Xₙ are iid N(μ, σ²) with σ known. For H₀: μ = μ₀ against H₁: μ = μ₁ with μ₁ > μ₀, the most powerful test rejects when:

  1. x̄ > μ₀ + z_α σ/√n
  2. x̄ > μ₁ + z_α σ/√n
  3. |x̄ − μ₀| > z_(α/2) σ/√n
  4. x̄ < μ₀ − z_α σ/√n
Answer and working

A. x̄ > μ₀ + z_α σ/√n

log Λ is increasing in x̄ when μ₁ > μ₀, so reject for large x̄, with the critical value set under H₀ to give size α. The critical value does not involve μ₁, which is why the same test is UMP for all μ₁ > μ₀.

Question 6 · easy

Counts in 6 classes are tested for fit to a model whose 2 parameters are estimated from the data by maximum likelihood. How many degrees of freedom does the G² (or Pearson χ²) statistic have?

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

D. 3

df = classes − 1 − parameters estimated = 6 − 1 − 2 = 3. 5 forgets the estimated parameters; 4 counts only one of them.

Question 7 · medium

Six farms compare a new and an old practice on adjacent strips; the yield differences (new − old, t/ha) are 0.4, 0.6, 0.2, 0.5, 0.3 and 0.4. What is the paired t-statistic and its degrees of freedom?

  1. t = 2.83 with 5 df
  2. t = 16.97 with 5 df
  3. t = 6.93 with 10 df
  4. t = 6.93 with 5 df
Answer and working

D. t = 6.93 with 5 df

Mean difference 0.40, SD 0.141, t = 0.40/(0.141/√6) = 6.93 with n − 1 = 5 df. 10 df is the unpaired pooled value; 2.83 forgets √n.

Question 8 · easy

Two varieties are compared on 12 and 10 plots with a pooled-variance t-test. How many degrees of freedom does the test have?

  1. 11
  2. 20
  3. 21
  4. 22
Answer and working

B. 20

n₁ + n₂ − 2 = 12 + 10 − 2 = 20.

Question 9 · medium

In a 2 × 2 table one expected frequency is 3. What is the recommended analysis?

  1. Fisher's exact test, because the χ² approximation fails with small expected counts
  2. Pooling the table with another experiment until every cell exceeds 3
  3. The z-test for two proportions, because it never needs large expected counts in any cell
  4. Pearson's χ² without correction, because small counts make it more powerful
Answer and working

A. Fisher's exact test, because the χ² approximation fails with small expected counts

With expected counts below 5, the χ² approximation can be poor; Fisher's exact test (conditioning on the margins) is the standard alternative. Yates' correction is a cruder fix. The z-test for two proportions is equivalent to the uncorrected χ² and has the same problem.

Question 10 · hard

For a single observation X from N(θ, 1), the most powerful level-α test of H0: θ = 0 against H1: θ = 1 rejects H0 when:

  1. |X| > z(α/2)
  2. X < −z(α)
  3. X > z(α)
  4. X² > χ²(α, 1)
Answer and working

C. X > z(α)

The likelihood ratio f1(x)/f0(x) = exp(x − 1/2) is increasing in x, so by the Neyman-Pearson lemma the most powerful test rejects for large x: X > z(α). The two-sided and chi-square forms are for a two-sided alternative and are not most powerful against θ = 1.

A study plan for Statistical Inference II: Testing of Hypotheses & Sequential Analysis in ICAR AIEEA-PG

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 (69% of the 35 questions): Tests based on Z, t, chi-square and F distributions (12, 34%); Hypotheses, the two types of error, size and power (6, 17%); Power and sample-size calculation for common tests (6, 17%).

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

  2. Then the middle band: Likelihood ratio tests and Wilks' theorem (3, 9%); Multiple testing: family-wise error rate and false discovery rate (3, 9%); Neyman-Pearson lemma and most powerful tests (3, 9%).

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

  3. Then the lighter units: Relation between tests and confidence intervals (1, 3%); Wald and score (Rao) tests (1, 3%).

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

  4. Leave time for the hard ones: Neyman-Pearson lemma and most powerful tests (1 hard of 3).

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

  5. Bank the easy marks: Likelihood ratio tests and Wilks' theorem (3 easy of 3); Multiple testing: family-wise error rate and false discovery rate (3 easy of 3).

    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 — ICAR AIEEA-PG Statistical Inference II: Testing of Hypotheses & Sequential 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.