ECS3706/ASSIGNMENT 2/2026
Assignment 02 Semester 01
Econometrics
ECS3706
Semester 1
Department of Economics
This tutorial letter contains Assignment 2 Questions.
, Question 1: Steps Involved in Performing Regression
Analysis
The following six steps are applied in regression analysis:
1. Review the literature and develop the theoretical model – Review existing research
and economic theory to understand the relationships between variables and to develop
a conceptual framework for the model.
2. Specify the model – Determine and define the dependent variable (Y) and
independent variables (X) based on economic theory. Choose the appropriate functional
form (e.g., linear, log-linear).
3. Hypothesize the expected signs of the coefficients – Based on economic theory,
postulate the expected direction of the relationship between each independent variable
and the dependent variable (positive or negative).
4. Collect the data, inspect and clean the data – Gather measurable data for the
variables specified in the model. Ensure data quality by checking for outliers, missing
values, and measurement errors.
5. Estimate and evaluate the equation – Use statistical software to estimate the
regression equation (typically using Ordinary Least Squares). Evaluate the results by
checking statistical significance (t-tests), overall fit (R²), and ensuring that the model
meets the underlying assumptions.
6. Document the results – Present the findings clearly, including the estimated
regression equation, interpretations of coefficients, and any limitations or challenges
encountered.
Question A2
(a) Explanation of the Residual Sum of Squares (RSS)
The Residual Sum of Squares (RSS), also known as the Error Sum of Squares (SSE), is the
sum of the squared differences between the observed values of the dependent variable
(Yᵢ) and the predicted values from the regression model (Ŷᵢ).
Mathematically: RSS = Σ(Yᵢ − Ŷᵢ)²
RSS measures the variation in the dependent variable that is not explained by the
regression model. It represents the portion of the total variation that remains after
accounting for the explanatory variables. A smaller RSS indicates that the model fits the
data better, as the predicted values are closer to the actual observed values.
The Ordinary Least Squares (OLS) method works by selecting coefficient estimates
that minimise the Residual Sum of Squares, thereby producing the best-fitting line
through the data points.
(b) Difference Between an Estimated Regression Line and a True
Population Regression Line
Assignment 02 Semester 01
Econometrics
ECS3706
Semester 1
Department of Economics
This tutorial letter contains Assignment 2 Questions.
, Question 1: Steps Involved in Performing Regression
Analysis
The following six steps are applied in regression analysis:
1. Review the literature and develop the theoretical model – Review existing research
and economic theory to understand the relationships between variables and to develop
a conceptual framework for the model.
2. Specify the model – Determine and define the dependent variable (Y) and
independent variables (X) based on economic theory. Choose the appropriate functional
form (e.g., linear, log-linear).
3. Hypothesize the expected signs of the coefficients – Based on economic theory,
postulate the expected direction of the relationship between each independent variable
and the dependent variable (positive or negative).
4. Collect the data, inspect and clean the data – Gather measurable data for the
variables specified in the model. Ensure data quality by checking for outliers, missing
values, and measurement errors.
5. Estimate and evaluate the equation – Use statistical software to estimate the
regression equation (typically using Ordinary Least Squares). Evaluate the results by
checking statistical significance (t-tests), overall fit (R²), and ensuring that the model
meets the underlying assumptions.
6. Document the results – Present the findings clearly, including the estimated
regression equation, interpretations of coefficients, and any limitations or challenges
encountered.
Question A2
(a) Explanation of the Residual Sum of Squares (RSS)
The Residual Sum of Squares (RSS), also known as the Error Sum of Squares (SSE), is the
sum of the squared differences between the observed values of the dependent variable
(Yᵢ) and the predicted values from the regression model (Ŷᵢ).
Mathematically: RSS = Σ(Yᵢ − Ŷᵢ)²
RSS measures the variation in the dependent variable that is not explained by the
regression model. It represents the portion of the total variation that remains after
accounting for the explanatory variables. A smaller RSS indicates that the model fits the
data better, as the predicted values are closer to the actual observed values.
The Ordinary Least Squares (OLS) method works by selecting coefficient estimates
that minimise the Residual Sum of Squares, thereby producing the best-fitting line
through the data points.
(b) Difference Between an Estimated Regression Line and a True
Population Regression Line
