4.13 Non-linear regression

4.13 Non-linear regression Flashcards

IB Maths: Applications and Interpretation HL: Flashcards

Studying this course? Keep your progress in one place.

Flip each card. Be honest: the ones you mark Not yet come back first.

14 cards

Tap a card to flip it. Swipe right if you knew it, left if not.

See all 14 cards
What does a least squares regression curve minimise?
The sum of the squares of the residuals.
What is a residual?
The observed value minus the value predicted by the model.
What does SSresSS_{res} measure?
The sum of squared residuals; a smaller value means a closer fit of the model.
What does R2R^2 tell you?
The proportion of the variability in the second variable accounted for by the model.
What is the relationship between R2R^2 and rr for a linear model?
R2=r2R^2=r^2.
Is R2R^2 alone a good way to choose between models?
No; consider context, extrapolation, complexity and whether the model makes sense.
Form of an exponential model?
y=abxy=ab^x; successive values at equal steps form a geometric sequence with ratio bb.
Form of a power model?
y=axby=ax^b.
Form of a sine model, and its period?
y=asin⁡(bx+c)+dy=a\sin(bx+c)+d; period 2πb\frac{2\pi}{b} with xx in radians.
What is the amplitude of y=asin⁡(bx+c)+dy=a\sin(bx+c)+d?
∣a∣|a|.
In N=3.50×1.38tN=3.50\times1.38^t, what does 1.381.38 mean?
The quantity grows by a factor of 1.381.38 per unit, i.e. 38%38\% growth.
Why is extrapolation risky?
The model is only supported by data in a given range and may not hold outside it.
A model has R2=0.95R^2=0.95. How should you interpret this?
95%95\% of the variability in the second variable is accounted for by the model.
What types of regression can be examined?
Linear, quadratic, cubic, exponential, power and sine.

Flashcards