Group A (Marks: 30)
(Related Course - STAT-305: Time Series Analysis)
This course is designed to explore the application of a variety of statistical models for time series in practical life using statistical software EViews and R.
Course Objectives:
The objectives of this course are to:
- Learn how to enter data, define variables, and perform variable manipulation and transformation as well as analyze the data. Specifically, reading and writing in R and EViews and other file types with different formats; Survey coding and data entry; selected data management procedures; and Data analysis and interpretation with R and EViews.
- Understand and be able to apply the concepts and methods underlying the analysis of univariate time series, and the context for interpretation of results.
- Decompose a time series into trend, seasonal, cyclical, and irregular components.
- Use various time series models such as moving average (MA), weighted MA, exponential smoothing, AR, ARMA, ARIMA, SARIMA, etc. for forecasting univariate time series data.
- Use multivariate time-series models such as vector auto regression (VAR) to analyze time series data.
- Forecast different time series.
Learning Outcomes:
After completing this course, students will be able to:
- Decompose a time series into trend, seasonal, and irregular components.
- Check stationarity and white noise of different time series.
- Test different time series.
- Calculate ACF and PACF.
- Identify statistical models and techniques that are appropriate for a particular type of time series data.
- Estimate and conduct inferences with time series models.
- Test the goodness of fit and adequacy of the identified time series models.
- Forecast the time series using appropriate time series models.
- Solve above time series oriented econometric analysis using R and EViews.
Contents:
- Reading and managing time series data, construction of time series plot, decomposition, forecasting by exponential smoothing, double exponential smoothing and Holt-Winter methods.
- Evaluation of ACVF and ACF, checking stationarity by different techniques including examination of ACVF and ACF, DF, ADF, Phillips-Perron tests for unit root, and Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test for trend stationarity.
- Estimation of AR, MA, ARMA, and ARIMA models and forecasting.
- Analyzing goodness of fit of the models and identifying the best model, simulating the models AR, MA, ARMA, and ARIMA and estimating parameters for forecasts.
- Estimating and forecasting by VAR model.
Rationale:
Group B (Marks: 20)
(Related Course - STAT-302: Sampling Technique-II)
This course is designed to provide a practical knowledge of drawing a sampling with appropriate sampling design and estimating parameters.
Course Objectives:
The objectives of this course are to:
- Understand using each sampling design in a particular situation.
- Draw samples using different sampling schemes.
- Estimate the parameters.
- Check the precision and calculate the errors.
- Interpret the results.
Learning Outcomes:
After completing this course, students will be able to:
- Apply particular sampling designs in real-life problems.
- Determine sample size for different sampling techniques and estimate mean, variance, proportions and construct confidence intervals.
- Design surveys using different sampling strategies, calculating estimates and assessing the precision of estimators.
- Conduct surveys using cluster sampling of unequal size and varying probability.
- Assess relative efficiency of cluster sampling compared with other sampling schemes.
- Conduct two-stage and multistage cluster sampling and obtain estimators with greater precision.
- Design double sampling procedures (stratified sampling, ratio estimator, regression estimator, product, PPS) and repetitive surveys (multiphase sampling).
- Understand the meaning of non-response error, characteristics of non-response error, and remedial measures of non-response error.
- Perform estimation procedures in the presence of non-response error.
- Perform inverse sampling, capture recapture method, network sampling, and snowball sampling schemes in particular problems.
Contents:
- Drawing probability samples, sampling with and without replacement.
- Estimation of population characteristics and variance of estimators for cluster sampling, double sampling, and two-stage sampling methods.
- Allocation of sample sizes for optimum cost and variance functions for different sampling procedures.
- Drawing stratified two-stage sampling and estimation of parameters.
- Related precision of different sampling schemes.