DP Math AI · HL · Statistics and Probability · Updated

AHL 4.13—Non-linear regression

Get started
Free preview 2/16
  1. Question 1

    A scatter plot of data shows increasing slowly at first, then increasing at a faster and faster rate as increases, with no turning points. Which model type is most appropriate?
    No clue? Show me the answer
    Correct answerCorrect!Incorrect
    AExponential

    Step-by-step walkthrough

    Choose a solution method

    Method #1Worked solution

    Step 1: Shape of data

    The data increases at an accelerating rate with no maxima, minima, or oscillation.

    Step 2: Match to model family

    An accelerating, ever-increasing growth pattern with no bends is the signature of exponential growth, .

    Step 3: Choose model

    The exponential model is the correct choice.

    Method #2Why the others are wrong

    Step 1: Quadratic

    A quadratic has a single turning point (a maximum or minimum), which is not present here.

    Step 2: Sine

    A sine model oscillates periodically around a mean, but this data never decreases.

    Step 3: Cubic

    A cubic model has an inflection with a change in curvature or up to two turning points, not a smooth accelerating increase.

    Step 4: Confirm

    Only the exponential model matches accelerating, non-oscillating, non-turning growth.

  2. Question 2

    A biologist models the relationship between an animal's body mass (kg) and its metabolic rate (watts) and finds the data forms a curve consistent with a proportional scaling law of the form . Which regression type should be used?
    No clue? Show me the answer
    Correct answerCorrect!Incorrect
    BPower regression

    Step-by-step walkthrough

    Choose a solution method

    Method #1Worked solution

    Step 1: Given model form

    The context explicitly states the relationship follows , a proportional scaling law.

    Step 2: Match to table of models

    This equation form corresponds to the power model, used for physical scaling laws such as mass versus metabolic rate.

    Step 3: Conclusion

    Power regression is the correct technique to fit .

    Method #2Why the others are wrong

    Step 1: Exponential

    Exponential regression fits , where is in the exponent, not raised as a base to a power.

    Step 2: Quadratic

    Quadratic regression fits , a fixed power of 2, not a variable exponent .

    Step 3: Sine

    Sine regression is for periodic data, not scaling relationships.

    Step 4: Confirm

    Only power regression directly matches the stated model form .

Free preview

14 more questions in this topic

Checking your topic access