Chapter 12 Flashcards
welke variabele op y as
y = reponse
welke variable op x as
x = explanatory
correlation falls between
-1 and 1
the stronger the linear correlation, (r)
the larger r is
where does the value of r NOT depend on
- the variables’ unit
- which one is the response and the explanatory variable
when is a value an outlier
- way higher or lower than other variables
- does not fall into the straight line of the regression
a subject who is average on x is predicted
to be average on y
oke
least squares line
de regressielijn, want sum of squares is het kleinste
The probability distribution of y values at a fixed value of x is a
conditional distribution
A regression model describes how the population mean my of each conditional distribution for the response variable depends on the value x of the explanatory variable. A straight-line regression model uses the line my = a + bx to connect the means. The model also has a parameter s that describes variability of observations around the mean of y at each x value.
oke
assumptions linear regression
- linear
- normal
- randomized
two sided significance test about population slope B
H0: b = 0, Ha: b =/= 0
test statistic: t = b-0/se
2* t.dist with df= n-2
95% CI for slope estimate
b +- t.0251*se
df = n-2
two sided test….
helft van p!!
interpret what the confidence interval implies…
On average, the response variable increases between (CI, CI) for every unit of the explanatory variable.