line of best fit ** Answ** doesn't mean it has to go through all points of the data, it just gets
the middle!
y hat is predicted and y is the actual values on graph
- line tries to minimize errors
sample correlation ** Answ** R
sum of squared errors ** Answ** what is being minimized on line of best fit
parameter ** Answ** the true information we want to understand about the population (mu)
correlation conditions ** Answ** four!
QQ, straight enough, no outliers, plot doesn't thicken (model's predicted power should be same
throughout)
residuals (e) ** Answ** the difference between the actual value (y) and its associated
predicted value (y^) so "actual - predicted"
meaning of a negative residual ** Answ** below the predicted value
How well does any line fit the data? ** Answ** calculate all the residuals and hope it sums
to 0
slope in the regression equation ** Answ** (b-sub1): for each one unit increase in X, we
expect Y to increase/decrease by b-sub1 on average
intercept in the regression equation ** Answ** (b-sub0)- when X is equal to zero, we expect
Y to equal b-sub0 on average
, intercepts can be meaningless when.. ** Answ** x is 0 and does not make sense
the predicted value of y does not make sense
R^2 ** Answ** NOT THE SAME AS R!!!
the % of variation in Y that is EXplained by the variation in X
statistical significance ** Answ** determined by the p value, <0.05 means not random,
significant; >0.05 means random, likely by random chance and insignificant
Two events that can easily be added together are ** Answ** mutually exclusive/disjoint
independent ** Answ** knowing one thing/event will not influence the next event; wo
events that have no influence on the probability of each other
- "and" determines multiplication
disjoint ** Answ** two things/events cannot occur at the same time/together, have no
outcomes in common
- used with word "or" and can be added
trial ** Answ** each occasion upon which we observe a random phenomenon
outcome ** Answ** the value of the random phenomenon noted at each trial
event ** Answ** result of combining two outcomes
sample space ** Answ** the collection of allll possible outcomes
law of large numbers ** Answ** in the long run, the frequency of repeated independent
events gets closer and closer to a single value, referring to the trials