ETF2100 Chap.2 Simple Regression and the OLS Fit
Simple Regression and the OLS Fit
Start from the observed condition
Simple regression represents a conditional mean with a straight line and leaves the remaining variation in an error term. The source review gives the linear relation y = β0 + β1x and the change relation Δy = β1Δx. This makes units central: if y is weekly earnings and x is years of schooling, the slope is dollars per week per additional year within the observed range.
OLS chooses a line from the sample, but the fitted slope inherits the quality of the data and the model. A neat line is not evidence that omitted determinants are unrelated to x.
The chapter objective is to read a fitted line as an intercept, slope and residual pattern rather than as a causal verdict. Begin by defining intercept at the scale used in the question.
Record whom or what intercept describes, its period or operating state, and evidence that distinguishes intercept from residual. Without that discipline, intercept can quietly change meaning between the opening claim and the final recommendation.
Next, make slope do explanatory work. State the direction of slope, the process it carries and the condition that keeps its link with intercept credible.
A useful slope note does not merely say that the relationship matters. It identifies which observation establishes intercept, which observation tests slope and which value of residual would force a different account.
Use residual as the chapter's discriminating lens. Compare at least two feasible cases and decide whether residual strengthens, narrows or reverses the preferred result.
If it cannot alter any conclusion, it is functioning as decoration. Attach the comparison to the same unit, population or system boundary used for intercept and slope.
Build the chapter explanation
A complete application of intercept has an actor, evidence, relationship and decision.
The actor has responsibility; evidence identifies the intercept state; slope explains why action may work; and residual supplies a review signal. This intercept–slope–residual structure makes ETF2100 reasoning auditable without turning one definition into a universal rule.
Take a fitted description wage = 420 + 55×education for workers whose education ranges from 10 to 18 years.
Moving from 12 to 13 years changes the fitted wage by 55 units because Δwage = 55×1. The intercept at zero years lies far outside the observed range and has little substantive meaning.
A worker with education 13 and observed wage 1,060 has a residual of 1,060 minus the fitted value 1,135, or −75. That residual is a model discrepancy, not proof of underpayment and not a direct observation of the population disturbance.
Now change one condition: Change education from years to a binary degree indicator. The one-unit slope now represents a group contrast, not an additional year.
Predict the direction of the result before consulting an example. Explain whether the change affects the definition of intercept, the mechanism carried by slope, the comparison represented by residual, or only the confidence attached to the conclusion.
Keep the controlling limit visible: A linear fit summarises the sample relation over its supported range; extrapolation and causal language need separate justification.
This residual limit is not ceremonial. It specifies the observation, design feature or operating condition that separates a careful use of intercept from a claim that outruns slope evidence.
For retrieval, close the explanation and reconstruct intercept, slope and residual in three different sentences: a definition, a relationship and a counter-case. Then attach one concrete ETF2100 example to each.
Reopen the residual material only to correct the first missing intercept–slope link; copying everything hides which analytical role failed.
For written or oral assessment, put the residual conclusion after the reasoning. Start with the requested decision, use intercept to establish the object and trace slope before allowing residual to challenge the preferred position.
Report residual at the scale earned by intercept evidence, preserving uncertainty and implementation constraints around slope.
Create an error log specific to intercept. Record the triggering fact, mistaken intercept inference, repaired relationship involving slope, and evidence from residual that distinguishes the two.
Repeat the repaired slope move on a different residual case so feedback becomes a transferable diagnostic for intercept.
A strong final check asks four questions. Is intercept defined consistently? Does slope explain a process rather than repeat the outcome? Can residual genuinely contradict the preferred answer?
Does the last sentence remain inside this limit: A linear fit summarises the sample relation over its supported range; extrapolation and causal language need separate justification. If any intercept–slope–residual answer is no, revise that defective relationship rather than adding more description.
What this chapter covers
- 01
intercept
- 02
slope
- 03
residual
- 04
read a fitted line as an intercept, slope and residual pattern rather than as a causal verdict
- 05
A linear fit summarises the sample relation over its supported range; extrapolation and causal language need separate justification.
Changed intercept case
- 1Define intercept at the required scale.
- 1Trace the role of slope.
- 1Use residual as a comparison or diagnostic.
- 1State the evidence that would change the conclusion.
- 1A linear fit summarises the sample relation over its supported range; extrapolation and causal language need separate justification.
Key terms
- intercept
- The fitted outcome when the explanatory variable equals zero.
- slope
- The fitted change in the outcome for a one-unit change in the explanatory variable.
- residual
- The observed outcome minus the value represented by the fitted relationship.
Simple Regression and the OLS Fit FAQ
How is intercept used in this chapter?
Define it at the task's unit and scale before applying slope.
What does slope explain?
It carries the relationship needed to read a fitted line as an intercept, slope and residual pattern rather than as a causal verdict.
Why does residual matter?
In Simple Regression and the OLS Fit, residual supplies a comparison, consequence or diagnostic capable of changing the conclusion.
What limits Simple Regression and the OLS Fit?
A linear fit summarises the sample relation over its supported range; extrapolation and causal language need separate justification.
Exam move
Retrieve intercept, slope and residual; explain their relationship; apply them to the changed case; then test the result against the stated boundary.
Working through Simple Regression and the OLS Fit in ETF2100? Sia is AskSia’s AI Econometrics tutor — ask any ETF2100 Simple Regression and the OLS Fit question and get a clear, step-by-step explanation grounded in how ETF2100 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
