review

Question 1

we come up with prediction in term of proabbility the use it to decide

Answer 1

if it 0 or 1 (categorical value)

Question 2

transformation is non linear so we use

Answer 2

odds which is ratio of probablity/ 1- probability

Question 3

in a regression outcome we are predicitng

Answer 3

log odds

Question 4

baseline goal is to predict

Answer 4

wheter observation will be 0 or 1

Question 5

baseline predicts

Answer 5

msot frequent outcome

Question 6

which data set to use to find outcome of baseline model

Answer 6

training

Question 7

how to build regression tree

Answer 7

splitting IV and predict most frequent outcome

Question 8

how to come up with prediction

Answer 8

count the nu,bers of outcome per split

Question 9

we choose how to define splits but use it

Answer 9

consisntley throught the model

Question 10

how to decide where to split

Answer 10

First decide what objective (error points that misclassified) to minimize and maximize accuracy, try different points and select one that minimize error or max accuracy

Question 11

most cases arent exact algorithim but

Answer 11

best found tree not optimal

Question 12

Annova class is for prediciting variable

Answer 12

limitless where as classification is for probability between 0 and 1

Question 13

continous is defined wiwthout a threshold how

Answer 13

most frewunt or verages

Question 14

classfication problem deals with probability either 0 or 1 and when use probabilty always have

Answer 14

threshold, speceficty, senstivity –> ROC curve

Question 15

single regression trees has high variance so prediciatbility will have

Answer 15

high variablity

Question 16

how to fix high variablity issue

Answer 16

multiple treees in random forest then make prediciton based on most frequent outcome from multiple trees

Question 17

• Build 5 trees and once have predictions from these trees then make prediction based on most frequent outcome-> can use idea of thresholds too how

Answer 17

if 3/5 trees = 1 then p y =1 is 0.6 then use this number to compare to thresholds
 If threshold is 0.7 predict 0 because 0.6<= 0.7

Question 18

we randomly selct what when build many trees

Answer 18

subset of variables. o Uses random row and column to make random and new data frame and based off this it is used to build trees and then repeat for multiple trees to make prediction on outcome

Question 19

N tree is

Answer 19

parameter usually between 200-500

Question 20

adjusting parameter doesnt really help becasue

Answer 20

default is good so just run algorithim

Question 21

random forest is generalization of regression trees and always perforrms betetr but disaadvantage is

Answer 21

less interpratble