Chapter 2 - Descriptive Analytics I

Question 1

What are the Characteristics that Define the Readiness Level of Data for an Analytics Study? (1-5)

Answer 1

Data source reliability - appropriateness of the medium where the data was obtained
Data content accuracy - data are correct and a good match for the analytics problem
Data accessibility - data are readily and easily obtainable
Data security and data privacy - only those who have authority and need to access the data can access it
Data richness - all the required data elements

Question 2

What are the Characteristics that Define the Readiness Level of Data for an Analytics Study? (6-10)

Answer 2

Data consistency - means data are accurately collected and combined/merged
Data currency/data timeliness - means the data should be up to date for a given analytics model
Data granularity - requires that the variables and data values be defined at lowest level of detail for the intended use of the data.
Data validity - the term used to describe a match/mismatch between the actual and expected data values of a given variable.
Data relevancy - means the variables in the data set are all relevant to the study being conducted.

Question 3

What has to match related to data and its usability?

Answer 3

The data has to match with the task for which it is intended to be used

Question 4

What does it mean to have data “analytics ready”?

Answer 4

It means the data has been transformed into a flat-file format and is ready for ingestion into predictive algorithms.

Question 5

What are the three broad sources of data in which business analytics come from?

Answer 5

Unstructured data, categorical structured data, and numerical structured data

Question 6

How is datum defined?

Answer 6

A singular form of data. A collection of facts usually obtained as the result of experiments, observations, transactions, or experiences.

Question 7

What is unstructured data?

Answer 7

Data composed of any combination of textual, imagery, voice, and Web content.

Question 8

What is structured data?

Answer 8

Categorical or numeric data that is used in data mining algorithms.

Question 9

What are the six elements of the data taxonomy?

Answer 9

Categorical Data
Nominal Data
Ordinal Data
Numeric Data
Interval Data
Ratio Data

Question 10

What are the two main categories of structured data and the two subcategories below them?

Answer 10

Structured data is either categorical or numerical.
Categorical data is either nominal or ordinal
Numerical data is either interval or ratio

Question 11

What are the subcategories for unstructured data?

Answer 11

Textual
Multimedia
XML/JSON

Question 12

What are the four steps of data preprocessing?

Answer 12

Data consolidation
Data cleaning
Data transformation
Data reduction

Question 13

What is dimensional reduction or variable selection as it relates to data preparation?

Answer 13

The reduction of variables that describe the phenomenon from different perspectives down to a manageable size

Question 14

What are the main tasks and methods for Data Consolidation?

Answer 14

Tasks: Access and collect the data, select and filter the data
Methods: SQL Queries, domain expertise, software agents

Question 15

What are the main tasks and methods for Data Cleaning?

Answer 15

Tasks: Handle missing values, ID and reduce noise, Find and eliminate errors
Methods: Fill in missing values, ID outliers and either remove or smooth the them, ID erroneous data

Question 16

What are the main tasks and methods for Data Transformation?

Answer 16

Tasks: Normalize, aggregate data; construct new attributes
Methods: Reduce range of values to a standard range; if needed, convert numeric variables to discrete variables; Derive new and more informative variables from existing ones

Question 17

What are the main tasks and methods for Data Reduction?

Answer 17

Tasks: Reduce number of attributes and records, balance skewed data
Methods: Principle component analysis; sampling; oversample the less represented or undersample the overrepresented classes.

Question 18

How is dispersion defined?

Answer 18

It is the representation of the numerical spread of a given data set.

Question 19

What is the box and whiskers plot?

Answer 19

A graphical illustration of both centrality and dispersion of a given data set

Question 20

What is correlation vs. regression?

Answer 20

Correlation is interested in the low-level relationships between two variables, regression is concerned with the relationships between all explanatory variables and the response variable.

Question 21

What is simple versus multiple regression?

Answer 21

A simple regression is built between one response variable and one explanatory variable. Multiple regression is built between one response variable and multiple explanatory variables.

Question 22

What are the three elements of developing regression models? (Revisit)

Answer 22

R squared
Overall F-test
Root mean square error (RMSE)

Question 23

What are the five assumptions to linear regression?

Answer 23

1. Linearity
2. Independence
3. Normality
4. Constant variance
5. Multicollinearity

Question 24

What is time series forecasting?

Answer 24

The use of mathematical modeling to predict future values of the variable of interest based on previously observed values.

Question 25

What are the three types of business reporting and what do each report?

Answer 25

Metric management reports - outcome-oriented metrics (e.g. KPIs)
Dashboard Reports - A range of different performance indicators on one page
Balanced-Scorecard Reports - Presents an integrated view of success in the org. From financial, customer, business process, and learning and growth perspectives.

Question 26

What is data visualization?

Answer 26

The use of visual representations to explore, make sense of, and communicate data.