LAISHA
Open Laisha

ICAR-IASRI MSc and PhD Causal Inference, Panel Data & Efficiency Analysis: 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
94written to this exam in this subject
12syllabus 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.

Official site: iasri.icar.gov.in.

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.

Causal Inference, Panel Data & Efficiency Analysis in ICAR-IASRI MSc and PhD

Impact evaluation of agricultural schemes with matching, difference-in-differences and IV; stochastic frontier and DEA efficiency analysis of farms; quantile regression.

Did the scheme, technology or variety cause the change? Potential outcomes, matching and weighting, difference-in-differences, instrumental variables, regression discontinuity and synthetic control, panel data models, quantile regression, and stochastic frontier and DEA efficiency analysis. Classical econometrics (OLS violations, simultaneous equations) is in the Agribusiness track and is assumed here.

Our 94 ICAR-IASRI MSc and PhD questions in this subject: 26 easy, 48 medium, 20 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 ICAR-IASRI MSc and PhD and are not shown on any other exam's page.

Question 1 · easy

What is the propensity score?

  1. The share of treated units in the sample
  2. The probability of treatment given covariates
  3. The difference in outcomes between matched pairs
  4. The predicted outcome under treatment
Answer and working

B. The probability of treatment given covariates

e(X) = P(D = 1 | X); Rosenbaum and Rubin showed it balances observed covariates between treated and untreated units.

Question 2 · hard

A matched evaluation of a credit scheme shows excellent balance on all observed covariates. What bias can remain?

  1. No bias at all once balance is achieved
  2. Bias from imbalance in the observed covariates
  3. Bias from unobserved confounding factors
  4. Bias from using the propensity score as a weight
Answer and working

C. Bias from unobserved confounding factors

Matching relies on selection on observables (unobserved confounders such as managerial ability are not handled); unobserved factors affecting both take-up and outcome still bias the estimate. Sensitivity analysis gauges how strong they would need to be.

Question 3 · medium

What does imposing common support do in propensity-score matching?

  1. Removes the need for a caliper
  2. Drops treated units without matches
  3. Adds controls to the sample until sizes match
  4. Forces all propensity scores to equal 0.5
Answer and working

B. Drops treated units without matches

Treated units whose scores exceed the controls' range have no counterparts and are dropped; the estimate then refers to the remaining units, which should be reported.

Question 4 · medium

At what level should standard errors be clustered in a DiD study of a district-level scheme with yearly farm data?

  1. The calendar year only
  2. Standard errors need no clustering
  3. The district
  4. The individual farm-year observation
Answer and working

C. The district

Cluster at the level of treatment assignment to allow for serial and within-district correlation; Bertrand, Duflo and Mullainathan showed unclustered DiD errors are far too small.

Question 5 · medium

A lottery offering subsidised tractors raises mean farm income by 6,000 currency units and tractor ownership by 30 percentage points. What is the Wald IV estimate of the effect of tractor ownership?

  1. 1,800 currency units
  2. 200 currency units
  3. 6,000 currency units
  4. 20,000 currency units
Answer and working

D. 20,000 currency units

Wald = reduced form / first stage = 6,000/0.30 = 20,000. 1,800 multiplies; 6,000 is the reduced form (ITT).

Question 6 · medium

Under monotonicity, what does an IV estimate with a binary instrument recover when effects differ across farmers?

  1. The average effect for compliers, the LATE
  2. The effect for the farmers with the largest gains
  3. The average effect for always-takers
  4. The average effect for the whole population
Answer and working

A. The average effect for compliers, the LATE

Imbens and Angrist showed IV identifies the local average treatment effect for units whose treatment responds to the instrument.

Question 7 · medium

Distance to the nearest input dealer is used as an instrument for fertiliser use. Which threat to exclusion is most plausible?

  1. Distance also shapes prices and access
  2. Distance is too strongly related to fertiliser use
  3. Distance has no variation between farms
  4. Distance can be measured precisely
Answer and working

A. Distance also shapes prices and access

Remote farms may also face lower output prices, less credit and fewer services, which affect income directly, violating exclusion.

Question 8 · easy

Which DEA model assumes variable returns to scale?

  1. SFA
  2. Malmquist
  3. CCR
  4. BCC
Answer and working

D. BCC

Banker, Charnes and Cooper (BCC, 1984) allows VRS; Charnes, Cooper and Rhodes (CCR, 1978) assumes CRS. SFA is econometric; Malmquist is a productivity index.

Question 9 · medium

A farm has an input-oriented DEA score of 1 but a non-zero slack in fertiliser. What does this mean?

  1. It is fully efficient in every sense
  2. Its output must be increased
  3. Fertiliser could still be cut
  4. It uses too little fertiliser
Answer and working

C. Fertiliser could still be cut

A radial score of 1 with slack means it lies on a vertical or horizontal part of the frontier; it is not Pareto-Koopmans efficient.

Question 10 · hard

A DEA study of 15 farms uses 6 inputs and 4 outputs, and most farms score 1. What is the most likely reason?

  1. Constant returns to scale was assumed
  2. Too many variables relative to the number of farms
  3. DEA always scores most units at 1
  4. The farms are genuinely all efficient
Answer and working

B. Too many variables relative to the number of farms

With many dimensions, each farm can be best in some direction; a common rule of thumb asks for at least three times as many DMUs as inputs plus outputs.

A study plan for Causal Inference, Panel Data & Efficiency Analysis 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.

  1. First, the heavy units (37% of the 94 questions): Propensity score matching and weighting (9, 10%); Randomised evaluation in agriculture and its limits (9, 10%); Stochastic frontier production functions and technical efficiency (9, 10%); Data envelopment analysis (8, 9%).

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

  2. Then the middle band: Difference-in-differences and event studies (8, 9%); Dynamic panels (introduction to GMM) (8, 9%); Instrumental variables and local average treatment effects (8, 9%); Panel data: fixed and random effects and the Hausman test (8, 9%).

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

  3. Then the lighter units: Potential outcomes and the identification problem (7, 7%); Quantile regression (7, 7%); Regression discontinuity designs (7, 7%); Synthetic control (6, 6%).

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

  4. Leave time for the hard ones: Dynamic panels (introduction to GMM) (3 hard of 8), Potential outcomes and the identification problem (2 hard of 7), Data envelopment analysis (2 hard of 8).

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

  5. Bank the easy marks: Panel data: fixed and random effects and the Hausman test (5 easy of 8); Synthetic control (3 easy of 6).

    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-IASRI MSc and PhD Causal Inference, Panel Data & Efficiency 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.