Data Analytics Exam

big data is often described by the four V’s

Volume, Velocity, Veracity, and Variety

What is volume

refers to the size of data

what is velocity

refers to speed/frequency of data processing

what is variety

refers to number of different types

what is veracity

refers to the data quality

what is the IMPACT model

Identify the question

Master the data

Perform the test plan

Address and refine results

Communicate insights

Track outcomes

the IMPACT cycle specifically includes all except

data preparation

descriptive analytics

is asking what is happening using categorization and classification of information

diagnostic analytics

is asking why did it happen, diagnostics are used to monitor changes in data

Predictive analytics

Is asking what’s going to happen, used to assess the likelihood of future outcomes

Prescriptive analytics

Is asking what should happen, used to forecasts they create to make recommendations

A company’s Sales decreased last month. An analyst examines customer data to determine why sales decreased. What type of analytics is this?

Diagnostic analytics

which of the following is an example of unstructured data

an email message with text and attachments

which attribute is required to exist in each table of a relational database and serves as the “unique identifier” for each record in a table?

primary key

the metadata that describe each attribute in a database is which?

Data Dictionary

what is the relational database function of a Primary Key

Serves as a unique identifier in a database table

what is the relational database function of a Foreign Key

creates a relationship between two tables

what is the relational database function of a Composite Primary Key

two foreign keys from the tables that it is linking combine to make up a unique identifier

what is the relational database function of a descriptive attributes?

Describe each record with characteristics with actual business information

what is the relational database function of a relational database

a means of storing data to ensure data are complete, redundant, to help enforce business rules

the ETL process stands for

Extract, transform, and load data

the purpose of transforming data is

to validate the data for completeness and integrity

what is the obtain stage of the ETL process

Use SQL to extract data from the source database

what is the Clean stage of ETL process

removing headings or subtotals

what is Determine purpose stage of the ETL process

choose which database and specific data will be needed to address the accounting question

what is the validating stage of ETL process

compare the number of records extracted to the number of records in the source database

what is the clean stage of ETL process

make sure all data formats start with two capital letters. Fix inconsistencies

what is the load stage of the ETL process

input the data into the analysis tool

the company wants to calculate the average order value for customers who have actually made a purchase. customers who have never purchased anything should not be included

inner join

the marketing team wants to create a list of every customer, including customers who have never placed an order. they want to identify customers with no sales so they can send them a reengagement campaign

left join

the data quality team wants to audit every order and identify orders where the customerID does not exist in the customers table. they specifically want to find the order associated with C999

right join

the company wants to reconcile the customer information with the order information in another system. they want to identify all mismatches in both directions customers with no orders and orders with no valid customer record

full outer join

what is data analytics

is the process of evaluating data with the purpose of drawing conclusions to address business questions

what is big data

refers to datasets which are too large and complex to be analyzed traditionally

Predictive analytics

what are the expected stock returns to our investment in Facebook stock?

Diagnostic analytics

Our refunds seem to be high, are they fraudulent?

Predictive Analytics

what are the cash needs and projections over the next 3 months?

descriptive analytics

what were the total taxes paid the past 5 years?

Prescriptive Analytics

If we expect our Asian sales to increase where should we produce them?

Prescriptive Analytics

Should we ship by truck, rail, or air given the expected increase in fuel expenses?

Descriptive Analytics

Which product sold the most last month?

Mode

which statistic identifies the value that occurs most frequently in a dataset?

a retail company finds that customers spend an average of $100 per purchase, with a standard deviation of $20. A customer spends $60, what is the customers Z-Score?

-2

A manager wants to identify employees wholes sales performance is unusually low, which employees should the manager investigate first?

employee C: Z score = -2.5

Data profiling typically involves unstructured data

False

in which scenario is an unsupervised learning approach typically used?

Segmenting customers into distinct groups based on their purchasing behavior

co-occurrence grouping

Which topics or keywords frequently appear together in customer reviews, like mentions of “fast shipping” and “excellent service”?

Clustering

Segmenting customers into groups based on past purchase history

profiling

Analyzing suppliers profiles to identify which suppliers have longer shipping times

Similarity matching

A bank wants to detect suspicious transactions. They compare each new transactions made by the same customer.

A delivery company uses the number of deliveries to predict fuel costs, fuel cost = 500 + 3 x Deliveries

Fuel cost increases by $3 for every additional delivery

a retailer predicts units sold based on product price, units sold = 1200 - 20 x Price. If the price increases by $5, what would you expect to happen to predicted units sold

decrease by 100 units

decision boundaries

techniques used to mark the location of the split between one class or another

test data

a set of data used to assess the degree and strength of a predicted relationship established by the training data

underfitting

a modeling error when the derived model poorly fits a limited set of data points

training data

existing data that have been manually evaluated and assigned a class

decision tree

tool used to divide data into smaller groups

overfitting

a model error when the derived model too closely fits a limited set of data points

a student score 0 points on an exam, which type of measurement is the exam score?

ratio

a company wants to show what percentage of its total revenue comes from each business segment

pie chart

a manager wants to compare the total sales of five different products

bar chart

a company wants to visualize its total sales by region and product category

stacked bar chart

a manager wants to identify which months and regions have the highest and lowest sales using different colors

heat map

you have 20 product categories and want to quickly identify the largest and smallest categories, which chart would be most appropriate?

tree map

useful for showing quartiles medians and outliers

box and whisker plots

correlation between two variables

scatter plots

distribution of sales across states or countries

filled geographic maps

visualize the line of best fits

scatter plots

data trends for net income over the past eight quarters

bar charts

data trends for stock price over the past 5 years

line chart