Stats SA3

Question 1

What does the data say will happen?

Answer 1

Predictive Analytics

Question 2

What has happened or what is happening now?

Answer 2

Descriptive Analytics

Question 3

Why it happened?

Answer 3

Diagnostic Analytics

Question 4

What will likely happen?

Answer 4

Predictive Analytics

Question 5

Predictive Analytics Process Order

Answer 5

Project Design, Data Sampling, Data Exploration, Data Modification, Model Validation, Model Development

Question 6

Kickoff meeting
Understand modeling objective
Define acceptance criteria
Document data and deployment requirement

Answer 6

Project Design

Question 7

Data extraction
Apply filters and exclusions
Identify external data sources

Answer 7

Data Sampling

Question 8

Exploratory data analysis
Identify data dependencies and correlations
Identify trends or anomalies in the data

Answer 8

Data Exploration

Question 9

Data Cleaning
Data augmentation and transformation
Feature selection

Answer 9

Data Modification

Question 10

Model performance review
Feedback based on business knowledge and inputs from subject matter experts (SME’s)

Answer 10

Model Validation

Question 11

Apply different modeling techniques and select final methodology

Answer 11

Model Development

Question 12

Linear Regression Analysis Formula

Answer 12

y = 6x + a + ε

Question 13

Dependent Variable (Value to be predicted)

Answer 13

y

Question 14

Beta coefficient (Rate multiplied to X)

Answer 14

6

Question 15

Independent variable (Value driving prediction)

Answer 15

x

Question 16

Alpha intercept (Baseline figure for y)

Answer 16

α

Question 17

Error term (Balancing figure)

Answer 17

ε

Question 18

Reasons for Inclusion for the Error Term (1) :

Answer 18

To account for unexplained variability in the dependent variable for other relevant independent variables, which may not have been included in the model

Question 19

Reasons for Inclusion for the Error Term (2) :

Answer 19

To capture measurement error in both the dependent and independent variables

Question 20

You can have more than one predictor variable (x1 - xn)

Answer 20

Multiple Linear Regression

Question 21

You still need to investigate the model’s _______

Answer 21

goodness-of-fit

Question 22

You need to prove if your predictors are _______

Answer 22

significant

Question 23

The _________, R^2, is a goodness-of-fit measure

Answer 23

coefficient of multiple determination

Question 24

_____ is a figure of merit; the higher the ____, the better is the success of the model in explaining the variation in the response using the set of predictors

Answer 24

R^2

Question 25

R^2 is normally expressed as a percentage and is interpreted as the amount of _____ in the response explained by the independent variables

Answer 25

variability

Question 26

is a decomposition of the total variation in the response into explained (pattern) and unexplained (error) parts

Answer 26

Analysis of Variance (ANOVA)

Question 27

The ______ variability is the amount of variation in the response variable that may be attributed to the predictors explicitly state in the model

Answer 27

explained

Question 28

The ______ variability is the amount of variation attribute to random error

Answer 28

unexplained

Question 29

SS refers to _____

Answer 29

Sum of Squares

Question 30

The df column refers to the _____

Answer 30

degrees of freedom

Question 31

The df for _____ is always the number of regression parameters minus one

Answer 31

Regression

Question 32

The df for ______, it is the sample size minus the number of regression parameters

Answer 32

Residual

Question 33

The total df is the ___ of those two degrees of freedom

Answer 33

sum

Question 34

MS refers to _____

Answer 34

Mean Squares

Question 35

The values in this column are the ratio of each sum of square to their respective degrees of freedom

Answer 35

Mean Squares

Question 36

_______ have no physical meaning but are instrumental in computing the F-statistic

Answer 36

Mean squares

Question 37

The _____ determines if regression is meaningful for the data at hand

Answer 37

F-test

Question 38

When the ____ is small. it means that there is at least one significant predictor in the analysis

Answer 38

p-value

Question 39

When _ is low, __ must go

Answer 39

p, Ho

Question 40

The p-value is low if it is less than the ____

Answer 40

a significance level

Question 41

The _____ helps in assessing if an individual predictor is significant

Answer 41

t-test

Question 42

In t-test, if p < 0.05, ____

Answer 42

Significant Predictor

Question 43

In t-test, if p > 0.05, ____

Answer 43

Insignificant Predictor