Stats 1 Flashcards
(11 cards)
Commenting on the likely reliability of an estimate using least squares regression line
- RELATIVE size of residuals
- How widely the points are scattered
- Therefore reliable/ not reliable
Why might x be a response variable?
X depends on Y
Residuals are found using:
Residual = y – (a+bx)
Mutually exclusive events (cannot happen at the same time):
P(A∩B) = 0 P(A∪B) = P(A) + P(B)
Independent events (do not affect each other):
P(A∩B) = P(A) x P(B) P(A/B) = P(A)
Exhaustive events (events where all possible outcomes are included):
P(A∪B) = 1
Why might the IQR be an appropriate measure of spread?
It is not affected by unknown/ extreme/ outlying values
Why might the standard deviation be an appropriate measure of spread?
Uses all data values
The value of r is unaffected by…
linear scaling
what is evidence that a set of data is modelled by a normal distribution?
- symmetrical
- not skewd
- bell-shaped
when revising comments…
talk about the change in
- residuals
- how more/less appropriate the model is
- stronger/weaker evidence of +ve relationship