Singapore University of Social Sciences

Selected Topics in Regression

Applications Open: To be confirmed

Applications Close: To be confirmed

Next Available Intake: To be confirmed

Schemes: To be confirmed

Language: English

Duration: 6 months

Fees: To be confirmed

Area of Interest: To be confirmed


Synopsis

ANL302e Selected Topics in Regression introduces students to the concepts and applications of regression. Its coverage includes selected topics in statistics, followed by advanced topics in multiple linear regression, panel data models, statistical forecasting models, and qualitative and limited dependent variable models. The course also includes analysing problems encountered in regression analysis where the standard assumptions are violated and look into methods to rectify these violations. The course aims to provide managers with the skills to perform regression analysis in the various business areas, such as estimation of demand, production and cost and business forecasting.

Level: 3
Credit Units: 5
Presentation Pattern: Every July

Topics

  • Multiple Regression Concepts
  • Use of Different Functional Forms
  • Further Inference in the Multiple Regression Model
  • Indicator Variables
  • Heteroscedasticity
  • Autocorrelation
  • Nonlinear Regression Models
  • Panel Data Models
  • Models with Binary Dependent Variables
  • Logit Model for Binary Choice
  • Basic Concepts in Business Forecasting
  • Business Forecasting Models

Learning Outcome

  • Describe the theoretical assumptions of regression models with reference to the Gauss Markov theorem.
  • Explain the strength and weaknesses of the various estimation methods in the regression analysis.
  • Formulate variants of the regression model with the various functional forms.
  • Test for problems arising from specification errors, heteroscedasticity, and autocorrelation.
  • Propose solutions to the abovementioned problems.
  • Apply business forecasting models, models with binary dependent variables and logit model for binary choice to appropriate data.
  • Interpret regression outputs produced by the modelling software.
Back to top
Back to top