Econometrics is often where the “rubber meets the road” for economics students. It is the sophisticated bridge that connects abstract economic theory with the messy, unpredictable reality of actual data. If economics tells us that raising prices should lower demand, econometrics asks: “By exactly how much, and how certain are we?” It is a discipline that requires a unique blend of economic intuition, linear algebra, and a healthy skepticism of “perfect” correlations.
Below is the exam paper download link
PDF Past Paper On Econometrics For Revision
Above is the exam paper download link
To help you navigate the world of regressions, residuals, and hypothesis testing, we have put together a high-impact revision guide based on the core challenges found in recent examination papers.
What is the “Ordinary Least Squares” (OLS) Method?
OLS is the workhorse of econometrics. Its goal is simple yet profound: to find the “Line of Best Fit” through a scatterplot of data. It does this by minimizing the sum of the squared differences between the actual data points and the values predicted by our model. In an exam, you will likely be asked to explain the Gauss-Markov Assumptions—the specific conditions (like linearity and no multicollinearity) that must be met for OLS to be the “Best Linear Unbiased Estimator” (BLUE).
How do we identify “Heteroskedasticity”?
In a perfect model, the “noise” (error term) should be consistent across all levels of your data. Heteroskedasticity occurs when the variance of these errors changes—for example, if the spending habits of wealthy families vary much more than those of low-income families. If ignored, it makes your standard errors unreliable. During revision, master the Goldfeld-Quandt and White tests, which are the standard “diagnostic tools” used to sniff out this problem.
What is “Multicollinearity” and why is it a problem?
Multicollinearity happens when two or more of your independent variables are too closely related to each other—like trying to measure the impact of both “Years of Education” and “Highest Grade Completed” on income. Because they move together, the model can’t figure out which variable is actually doing the work. This leads to high standard errors and “unstable” coefficients that change wildly with small bits of new data.
What is the purpose of the “Durbin-Watson” Statistic?
When dealing with Time Series data, we often run into Autocorrelation—where today’s error is linked to yesterday’s error. The Durbin-Watson statistic is a numerical “smoke detector” for this issue. A value near 2.0 suggests no autocorrelation, while values moving toward 0 or 4 indicate a problem that needs to be fixed using methods like the Cochrane-Orcutt procedure.
How do “Dummy Variables” work in a model?
Not all data is numerical. How do you measure the impact of “Gender,” “Location,” or “Policy Change”? You use Dummy Variables (0 or 1). A key trap to avoid in your finals is the Dummy Variable Trap—if you have two categories (Male/Female), you only include one dummy variable in the regression to avoid perfect multicollinearity with the intercept.
What is “Endogeneity” and the “Instrumental Variables” (IV) solution?
Endogeneity is the “nightmare” of econometrics; it happens when an independent variable is correlated with the error term, often because of a “hidden” third factor or reverse causality. To solve this, we use an Instrument—a variable that is related to our independent variable but has no direct connection to the error term. It’s like using a “proxy” to isolate the true effect you are trying to measure.
Conclusion
Econometrics is a “detective” science. It’s about looking at a set of results and asking: “Is this relationship real, or is it a statistical ghost?” Success in your exam comes from your ability to interpret the output—looking at the R-squared, the t-statistics, and the p-values to tell a story that makes economic sense.
To help you practice your model specifications and diagnostic testing, we have provided a link to the essential revision materials below.
Last updated on: March 24, 2026