week 5

Question 1

predictive analytics use historical data to

Answer 1

tell us somehting about the future

Question 2

IV vs DV

Answer 2

IV - used to predict DV and on the x axis
DV- what we trying to predict based on other variables on y axis

Question 3

what are we trying to predict with lm

Answer 3

B0, B1 or coeffcients

Question 4

what are we trying to minimze for each variable and for each model

Answer 4

E for variablke and total error for model

Question 5

Why use SSE

Answer 5

magnify error deviation by squaring it

Question 6

issue with SSE and fix

Answer 6

depedns on number of points- more points SSE = higher
fix- by use RMSE: normalized by N and same variable as DV so if DV is price units will be in $

Question 7

R square is high means

Answer 7

model fits well with data and error are small but not guarantee work well on unseen data

Question 8

MAPE vs MAE in high averages for data set

Answer 8

MAPE better if data set average high as will show as a %. MAE will be higher if data set higher

Question 9

what is r squared

Answer 9

is percentage decrease in SSE, what percentage in SSE has actually dropped compared to baseline model (SST)

Question 10

it is hard to get model with good accuracy 0.8+ on real data so what values good

Answer 10

0.3

Question 11

model r ^2 get better if you add more variables that are above 0 R^2 but at a

Answer 11

diminish rate

Question 12

• Not all variables should be used because

Answer 12

model over fit data

Question 13

issue with over fitting

Answer 13

is that it will perform badly on unseen data because it doesn’t know that data just memorized old data
 Will change coefficient to minimize error when given to make prediction on future will make error because is over fitted

Question 14

significance is based on confidence level we want if confidence is 95% pvalue is and what is insiginifcant

Answer 14

5% if greater then not signifacnt

Question 15

 Coefficient (beta) = 0.6 means

Answer 15

if IV increase by 1 unit then DV will increase by 0.6 units.

Question 16

sign of overfitting

Answer 16

Adjusted r square can increase or decrease. Add new variable and adjusted r square goes down = overfitting the model

Question 17

goal is to include only significant variables in regression because

Answer 17

other variables will cause overfitting

Question 18

correlation does what

Answer 18

mirrors linear relationship between two variables. It measures the degree to which the two variables are linearly related to each other is between -1 and 1

Question 19

linear regression assumption about correlation

Answer 19

all variables independant so no correlation

x variables are going to be independent not dependant on other variables

Question 20

whats is sign for worry in correlation

Answer 20

-0.6-0.6

Question 21

why do we split data

Answer 21

model may just be trying to minimize error not make predictions, to see if perform well or just overfit, training should be 80%, using lm function with training data to build

Question 22

if coeefcient 0 means

Answer 22

no impact on dv from iv

Question 23

output of RMSE/ MAE

Answer 23

tells us predictions within # error on average differnece is RMSE gives more weigh to larger errors. making senssitve to outliers or large deviations

Question 24

MAE used when

Answer 24

average magnitude of errors, regardless of directions. less sensitive to outliers with focus on overall accuracy not punish deviations

Question 25

MAPE output tells us

Answer 25

avarge deviation as a % from actual price

Question 26

r^2 only useful for who adn solution

Answer 26

analysts, use RMSE, MAPE, MAE as easeier to understand

Question 27

when r = 0

Answer 27

indicates that there is no linear correlation between x and y. However, it does not necessarily imply that there is no relationship between them.

Question 28

best metrics for testing data

Answer 28

MAE, RMSE, MAPE

Question 29

what does -R^2 mean

Answer 29

o This means baseline model (average of price = price all wines) is doing better than your model and yours is useless. So build better model or just use baseline

Question 30

what is standard error

Answer 30

meausre of uncertantiy in estimate of coeffcient

Question 31

what are residuals in model summary

Answer 31

• residuals are when you build a model, tell you the distribution of these errors