LAISHA
Open Laisha

Indian Statistical Service Examination Time Series Analysis: practice questions

Indian Statistical Service Examination (with Indian Economic Service) · India · Union Public Service Commission

Original practice questions written to the published syllabus · not past papers
10worked here, free
99written to this exam in this subject
12syllabus units

About Indian Statistical Service Examination

Indian Statistical Service Examination (with Indian Economic Service) is a government recruitment exam run by Union Public Service Commission, for candidates in India. 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.

Official site: upsc.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.

Time Series Analysis in Indian Statistical Service Examination

Modelling observations ordered in time, from classical decomposition to Box-Jenkins and beyond: stationarity, ARIMA and seasonal ARIMA, volatility models for prices, vector autoregressions, state-space models and the Kalman filter, and forecast evaluation. Agribusiness econometrics introduces ARIMA and unit roots; this subject is the full treatment.

Our 99 Indian Statistical Service Examination questions in this subject: 27 easy, 50 medium, 22 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 Indian Statistical Service Examination and are not shown on any other exam's page.

Question 1 · easy

Holt's linear method gives a current level of 120 tonnes and a current slope of 4 tonnes per month. What is the forecast three months ahead?

  1. 124 tonnes
  2. 136 tonnes
  3. 120 tonnes
  4. 132 tonnes
Answer and working

D. 132 tonnes

The h-step forecast is L + h b = 120 + 3 x 4 = 132. 124 is the one-step forecast; 120 ignores the trend, which is what simple exponential smoothing would give.

Question 2 · hard

What does the innovations state-space (ETS) framework add to exponential smoothing methods?

  1. A likelihood, so parameters, AIC and prediction intervals have a statistical basis
  2. A guarantee that every smoothing method equals some ARIMA model
  3. Removal of the need to choose smoothing constants at all
  4. Forecasts that never depend on the error distribution assumed
Answer and working

A. A likelihood, so parameters, AIC and prediction intervals have a statistical basis

ETS models give each smoothing method an underlying stochastic model with a single error source, allowing maximum likelihood, information criteria and proper prediction intervals. Multiplicative ETS models have no ARIMA equivalent, so the ARIMA claim is false; constants are still estimated.

Question 3 · easy

The sample ACF of a monthly price series starts near 0.95 and declines very slowly and almost linearly over 30 lags. What should the analyst do first?

  1. Difference the series
  2. Fit an MA(30)
  3. Fit an AR(1) model to the levels at once
  4. Conclude the series is white noise and stop
Answer and working

A. Difference the series

A slow, nearly linear decay is the signature of non-stationarity (a unit root); the series should be differenced (and tested formally) before identifying ARMA orders. An MA(30) is grossly over-parameterised, and the ACF clearly rules out white noise.

Question 4 · medium

Why is a random walk, Y(t) = Y(t-1) + e(t) with e(t) white noise, not stationary even though its expected change is zero?

  1. Its variance grows with time
  2. Its errors are correlated with one another
  3. Its mean rises in a straight line with time
  4. Its first difference is a non-stationary series
Answer and working

A. Its variance grows with time

Starting from Y(0), Var(Y(t)) = t sigma^2, which grows without limit, so the variance is not constant. The mean stays at Y(0) with no drift; the shocks are white noise; and the first difference is white noise, which is stationary.

Question 5 · easy

The process Y(t) = 4 + 0.6 Y(t-1) + e(t) is stationary. What is its mean?

  1. 4
  2. 10
  3. 2.4
  4. 6.67
Answer and working

B. 10

mu = c/(1 - phi) = 4/0.4 = 10. 4 is the constant, not the mean; 6.67 = 4/0.6; 2.4 = 4 x 0.6.

Question 6 · easy

What is the order of the stages in the Box-Jenkins approach to model building?

  1. Identification, estimation, diagnostic checking, then forecasting
  2. Forecasting, estimation, identification, then diagnostic checking
  3. Estimation, identification, forecasting, then diagnostic checking
  4. Diagnostic checking, identification, estimation, then forecasting
Answer and working

A. Identification, estimation, diagnostic checking, then forecasting

Box and Jenkins proposed an iterative cycle: identify a tentative model, estimate it, check the residuals, and return to identification if the checks fail; forecasting follows once the model is adequate.

Question 7 · hard

When is a regression on 11 monthly dummy variables with ARMA errors preferable to seasonal differencing?

  1. When the series has a seasonal unit root at lag 12
  2. When the seasonal pattern drifts and changes from year to year
  3. When the seasonal pattern is deterministic and stable over the years
  4. When fewer than two years of data are available
Answer and working

C. When the seasonal pattern is deterministic and stable over the years

Seasonal differencing assumes a stochastic seasonal pattern that can drift (a seasonal unit root); if seasonality is fixed, dummies are more efficient and avoid over-differencing. Tests such as HEGY and Canova-Hansen help decide.

Question 8 · medium

A regression of one district's wheat price on an unrelated district's cotton area, both trending random walks, gives R-squared = 0.92 and a Durbin-Watson statistic of 0.25. What is the most likely explanation?

  1. Perfect multicollinearity between the two regressors
  2. Heteroscedastic errors with no autocorrelation
  3. A genuine long-run relationship confirmed by the high R-squared
  4. A spurious regression between non-stationary series
Answer and working

D. A spurious regression between non-stationary series

Granger and Newbold showed that regressing independent random walks gives high R-squared and very low Durbin-Watson; R-squared above DW is a warning sign. The fit says nothing about a real relation unless the series are cointegrated.

Question 9 · easy

Two I(1) price series are said to be cointegrated when:

  1. Both become stationary after first differencing
  2. Some linear combination of them is stationary
  3. One of them Granger-causes the other in a VAR
  4. Their correlation coefficient exceeds 0.9 in levels
Answer and working

B. Some linear combination of them is stationary

Cointegration means a linear combination is I(0), so the series share a long-run equilibrium. Being I(1) individually is the precondition, not the definition; high correlation and Granger causality are neither necessary nor sufficient.

Question 10 · easy

How does the Kalman filter handle a month in which the market was closed and no price was recorded?

  1. It stops and requires the series to be restarted after the gap
  2. It replaces the missing value by the series mean before filtering
  3. It skips the update step and carries the prediction forward
  4. It deletes the month and joins the series together
Answer and working

C. It skips the update step and carries the prediction forward

With no observation, the filtered state equals the predicted state and its variance grows by the state noise. This handles gaps without imputation, a key advantage for agricultural series.

A study plan for Time Series Analysis in Indian Statistical Service Examination

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 (41% of the 99 questions): AR, MA and ARMA processes (11, 11%); Forecast accuracy: MAPE, RMSE and the Diebold-Mariano test (10, 10%); Unit-root tests and structural breaks (10, 10%); Vector autoregression and cointegration (10, 10%).

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

  2. Then the middle band: ARCH and GARCH models for price volatility (9, 9%); Box-Jenkins ARIMA: identification, estimation and diagnostic checking (9, 9%); Seasonal ARIMA and ARIMAX with weather covariates (9, 9%); Stationarity, autocorrelation and partial autocorrelation functions (9, 9%).

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

  3. Then the lighter units: State-space models and the Kalman filter (8, 8%); Spectral analysis (introduction) (7, 7%); Smoothing: moving averages and exponential smoothing (Holt-Winters) (6, 6%); Components of time series and classical decomposition (1, 1%).

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

  4. Leave time for the hard ones: Seasonal ARIMA and ARIMAX with weather covariates (3 hard of 9), Unit-root tests and structural breaks (3 hard of 10), Spectral analysis (introduction) (2 hard of 7).

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

  5. Bank the easy marks: Stationarity, autocorrelation and partial autocorrelation functions (4 easy of 9); Seasonal ARIMA and ARIMAX with weather covariates (3 easy of 9).

    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 — Indian Statistical Service Examination Time Series 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.