Indian Statistical Service Examination (with Indian Economic Service)
India · Union Public Service Commission · unverified source
851practice questions written to this syllabus
35subjects
What Indian Statistical Service Examination is
Indian Statistical Service Examination (with Indian Economic Service) is a government recruitment exam run by Union Public Service Commission, for candidates in India.
Format
Written examination of six papers and a viva voce. Reported scheme: General English 100, General Studies 100, Statistics I and II (objective, 200 each), Statistics III and IV (descriptive, 200 each); viva 200.
Statistics II: linear models, estimation, hypothesis testing, multivariate analysis
Statistics III: sampling techniques, design of experiments, economic statistics, SQC and operations research (reported)
Statistics IV: optional areas such as stochastic processes, demography and vital statistics, computing, time series and econometrics (reported)
General English; General Studies
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: upsc.gov.in.
How it has changed
ISS officers staff MoSPI, the National Statistical Office and statistics divisions of ministries including agriculture. Paper contents are as Laisha knows them and are marked 'reported' in curriculum weights.
How to prepare for Indian Statistical Service Examination
Where our 851 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.
Heaviest (72% of the questions): Time Series Analysis (99); Demography & Vital Statistics (98); Statistical Quality Control & Reliability (84); Causal Inference, Panel Data & Efficiency Analysis (59); Index Numbers, National Accounts & Agricultural Economic Statistics (58); Optimization Techniques for Statisticians (57); Probability Distributions & Sampling Distributions (32); Probability Theory (27); Statistical Inference II: Testing of Hypotheses & Sequential Analysis (27); Nonparametric & Rank-Based Methods (25); Sampling Theory I: Simple, Stratified, Systematic, Ratio & Regression Estimation (25); Statistical Inference I: Estimation (24).
Start here: read these subjects' manuals first and work their questions until the reasoning is automatic.
Middle band: Bayesian Inference (20); Crop Area & Yield Estimation: Crop-Cutting Experiments (20); Multivariate Analysis (19); Stochastic Processes (16); Survey Design, Questionnaires, Field Operations & Non-Sampling Errors (16); Agricultural Statistics System, Census & Official Statistics (15); Linear Models & Regression Analysis (13); Incomplete Block Designs & Design Construction (13); Statistical Methods for Agricultural Sciences (11); Principles of Experimental Design & Basic Designs (10); Differential Equations, Finite Differences & Numerical Analysis (9); Groups of Experiments, Multi-Location Trials & Genotype x Environment Analysis (9).
One pass per subject, then mixed sets across them.
Lighter: Specialised Agricultural Experiments: Cropping Systems, Animal & On-Farm Trials (8); Linear Mixed Models, Variance Components & Spatial Analysis of Field Trials (8); Advanced Design of Experiments (PhD) (8); Sampling Theory II: Unequal Probability, Cluster, Multistage & Repeated Sampling (8); Bioassay, Survival & Epidemiological Statistics (7); Small Area Estimation & Model-Based Survey Inference (6); Statistical Genetics: Population & Quantitative Genetics (6); Response Surface, Mixture & Optimal Designs (5); Biometrical Techniques in Plant & Animal Breeding (5); Categorical Data Analysis & Generalized Linear Models (3); Factorial Experiments, Confounding, Split- & Strip-Plot Designs (1).
Definitions, classifications and standard figures; quick to revise, easy marks to lose.
Hardest share: Bayesian Inference (11 hard of 20), Linear Models & Regression Analysis (7 hard of 13), Multivariate Analysis (9 hard of 19).
Leave untimed time for these and read every explanation, including for the options you rejected.
Original questions written to this exam's published syllabus and pattern. Not past papers.
1. A stable population has intrinsic birth rate 0.035 and intrinsic rate of natural increase 0.015. What is the intrinsic death rate?
0.020
0.050
0.015
0.035
Answer and working
A. 0.020
In a stable population d = b − r = 0.035 − 0.015 = 0.020.
Demography & Vital Statistics · Stable and stationary populations
2. What is the efficiency factor of a BIBD with v = 7, k = 3 (r = 3, lambda = 1)?
0.429
1.286
0.667
0.778
Answer and working
D. 0.778
E = lambda v/(rk) = 7/9 = 0.778, equal to v(k - 1)/[k(v - 1)] = 14/18.
Incomplete Block Designs & Design Construction · Efficiency factors of incomplete block designs
3. Which algorithm solves the assignment problem efficiently?
Vogel's approximation
The two-phase method
The Hungarian method
The north-west corner rule
Answer and working
C. The Hungarian method
The Hungarian method exploits the assignment structure; NW corner and Vogel give initial transportation solutions.
Optimization Techniques for Statisticians · Transportation and assignment problems
4. Three components with reliabilities 0.95, 0.98 and 0.99 are in series. What is the system reliability?
0.922
0.973
0.950
0.999990
Answer and working
A. 0.922
Series: product = 0.922. 0.999999 is the parallel arrangement.
Statistical Quality Control & Reliability · Reliability: hazard rate, series and parallel systems, life testing
5. A dual-record sample registration system combines continuous recording of births and deaths by a local enumerator with which second element?
Verbal autopsy interviews for every recorded birth
An independent periodic survey whose records are matched to the continuous ones
Hospital records of all births and deaths occurring anywhere in the whole country
Verbal autopsy interviews carried out for every birth recorded by the enumerator
Answer and working
B. An independent periodic survey whose records are matched to the continuous ones
Matching two independent sources allows estimation of events missed by both (as in capture-recapture), correcting for under-recording.
Demography & Vital Statistics · Sources of demographic data: census, civil registration and sample registration
6. How many conditions define the Moore-Penrose inverse uniquely?
Two
One
Four
Three
Answer and working
C. Four
AGA = A, GAG = G, (AG)' = AG and (GA)' = GA. The single condition AGA = A defines g-inverses in general.
Linear Models & Regression Analysis · Less-than-full-rank models and generalised inverses
7. A stationary series has rho(1) = 0.6 and rho(2) = 0.5. What is the partial autocorrelation at lag 2?
0.360
0.140
0.500
0.219
Answer and working
D. 0.219
phi(2,2) = (rho2 - rho1^2)/(1 - rho1^2) = (0.5 - 0.36)/0.64 = 0.219. Forgetting the denominator gives 0.140; 0.500 is the ordinary autocorrelation; 0.360 is rho1 squared, the lag-2 ACF an AR(1) would imply.
Time Series Analysis · Stationarity, autocorrelation and partial autocorrelation functions
8. For subgroups of 5 with R̄ = 0.40 kg, D₃ = 0 and D₄ = 2.114. What is the R chart upper limit?
0.846 kg
1.20 kg
0.231 kg
0.400 kg
Answer and working
A. 0.846 kg
UCL = D₄R̄ = 2.114 × 0.40 = 0.846; LCL = 0.
Statistical Quality Control & Reliability · Control charts for variables: X-bar, R and s charts