ICAR-IASRI MSc and PhD Crop Forecasting, Weather-Based Models & Non-linear Growth Models: 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
40written 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.
Crop Forecasting, Weather-Based Models & Non-linear Growth Models in ICAR-IASRI MSc and PhD
Developing and validating district-level forecast models, fitting logistic, Gompertz and Richards models with iterative methods, composite and remote-sensing models, and calibration and evaluation of crop simulation models.
Forecasting harvests before they happen: Fisher's regression integral and weather-indices models, pre-harvest forecasts from biometric characters and farmers' appraisal, agro-meteorological and remote-sensing models, non-linear growth curves (logistic, Gompertz, Richards), and how statistical and crop-simulation models are combined.
Our 40 ICAR-IASRI MSc and PhD questions in this subject: 6 easy, 16 medium, 18 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 · hard
How is a growth index constructed from biometric characters observed at several stages?
Characters are averaged with equal weights across all stages
The largest character value across stages is taken as the index
Weighted by their correlation with yield
Each character is divided by its value at the previous stage
Answer and working
C. Weighted by their correlation with yield
Characters at each stage are combined with weights equal to their correlation with yield, the same idea as weather indices, summarising crop growth to date in a few variables.
Question 2 · hard
Under which sampling method from survey theory is the calibration of eye estimates against crop-cutting yields formally analysed?
Sampling on successive occasions with full replacement
Double sampling with regression estimation
Cluster sampling with equal-sized clusters
Systematic sampling with a random start in each village
Answer and working
B. Double sampling with regression estimation
A large first-phase sample gives cheap appraisals and a small second-phase sample gives both appraisal and objective yield; the regression estimator of double sampling combines them.
Question 3 · hard
When does combining forecasts from weather, remote-sensing and biometric models reduce error most?
When all three models have exactly the same errors
When the sources carry different information and their errors are not highly correlated
When the combination weights are chosen to fit the test years
When one source is far worse than the other two
Answer and working
B. When the sources carry different information and their errors are not highly correlated
Combining averages out independent errors; if errors are highly correlated there is little gain, and choosing weights on the test years is leakage.
Question 4 · hard
Which index was designed to reduce the influence of soil background on vegetation indices in sparse canopies?
Simple Ratio of red to blue reflectance
Land Surface Temperature index
Normalised Difference Water Index (NDWI)
Soil-Adjusted Vegetation Index (SAVI)
Answer and working
D. Soil-Adjusted Vegetation Index (SAVI)
SAVI adds a soil-adjustment constant to the NDVI formula to reduce soil brightness effects where the canopy is sparse. NDWI targets water content; LST is thermal.
Question 5 · medium
In the Richards model Y = K(1 + b exp(-rt))^(-1/m), which special case is obtained with m = 1?
The logistic model
The Gompertz model
The monomolecular model
The exponential model
Answer and working
A. The logistic model
With m = 1 the formula is K/(1 + b exp(-rt)), the logistic. The Gompertz is the limit as m tends to 0; the monomolecular corresponds to m = -1 with b negative.
Question 6 · hard
Why is a fitted asymptote K unreliable when growth data cover only the early, nearly exponential phase?
Few data lie near the plateau
Early growth data always contain negative values
The rate parameter becomes exactly zero in early data
The asymptote can only be estimated by linear regression
Answer and working
A. Few data lie near the plateau
In the early phase several combinations of K and r fit almost equally well, so K has a huge standard error and is highly correlated with r. Extrapolating to final size is unreliable.
Question 7 · medium
Why is R-squared a weak criterion for comparing non-linear growth models?
It decreases whenever a parameter is added
It is always exactly 1 for non-linear models
It lacks the usual sum-of-squares decomposition
It can only be computed for linear models with two parameters
Answer and working
C. It lacks the usual sum-of-squares decomposition
For non-linear models the total sum of squares does not split into regression and residual parts as in linear regression; RMSE, residual plots and information criteria are preferred.
Question 8 · easy
What do crop simulation models such as the CERES models in DSSAT compute?
A single regression of yield on total rainfall
Daily crop growth from weather, soil, management and cultivar data
Only the market price of the crop at harvest
Area under the crop from satellite images alone
Answer and working
B. Daily crop growth from weather, soil, management and cultivar data
Process-based simulation models step through the season day by day, simulating phenology, biomass, partitioning, and soil water and nitrogen balances.
Question 9 · medium
A crop model's simulated yields have a correlation of 0.95 with observed yields across 20 sites but are on average 1.2 t/ha too high. What does this show?
Correlation and bias are the same measure
The observations must contain a 1.2 t/ha error
Correlation is high but the model is biased
The model is accurate because the correlation is above 0.9
Answer and working
C. Correlation is high but the model is biased
High correlation does not guarantee agreement: correlation measures association; a model can track variation well yet be systematically off. Mean bias, RMSE and the slope and intercept of observed on simulated values reveal it.
Question 10 · medium
A Nash-Sutcliffe efficiency of -0.3 for a crop model on validation data means:
The model is unbiased but imprecise
The model is better than the observed mean by 30 per cent
It is worse than the observed mean
The model explains 30 per cent of the variance
Answer and working
C. It is worse than the observed mean
NSE = 1 is perfect, 0 equals the skill of the observed mean, and negative values are worse than the mean.
A study plan for Crop Forecasting, Weather-Based Models & Non-linear Growth Models 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 (70% of the 40 questions): Crop simulation models and their statistical calibration (10, 25%); Machine-learning and hybrid forecasting models (9, 23%); Fitting non-linear models: iterative estimation and goodness of fit (6, 15%); Remote-sensing-based yield forecasting with vegetation indices (3, 8%).
Read the manual chapters for these first and do every worked question you can find on them.
Then the middle band: Weather-based models: Fisher's regression integral and weather indices (3, 8%); Farmers' eye estimates and appraisal-based forecasts (2, 5%); Models using plant biometric characters (2, 5%); Non-linear growth models: logistic, Gompertz, Monomolecular and Richards (2, 5%).
Aim for one sitting per unit, then mixed questions across them.
Then the lighter units: Agro-meteorological and composite models (1, 3%); Evaluation of forecasts and their use in policy (1, 3%); Pre-harvest forecasting: purpose and history (1, 3%).
Learn the definitions, classifications and standard numbers; these units are more often recalled than worked.
Leave time for the hard ones: Farmers' eye estimates and appraisal-based forecasts (2 hard of 2), Agro-meteorological and composite models (1 hard of 1), Evaluation of forecasts and their use in policy (1 hard of 1).
Come back to these after a first pass, untimed, and read every explanation, including for the options you rejected.
Bank the easy marks: Machine-learning and hybrid forecasting models (3 easy of 9); Crop simulation models and their statistical calibration (3 easy of 10).
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.