Business Analytics Exam 1

observation

row of data

multicollinearity

strong correlation between 2 independent variables

residual

observed minus predicted, error, what was gotten wrong

first step of data mining

Purpose

second step of data mining

Obtain

third step of data mining

Clean

fourth step of data mining

Build

fifth step of data mining

Evaluate

sixth step of data mining

Deploy

If p ≤ α

reject null hypothesis

Contingency tables

used to study relationship between two or more categorical variables

normality

means that leftover errors (residuals) will represent bell shaped curve centered at 0

independence

means that leftover error for one data point has no relationship to another

constant variance

messiness of leftover errors is represented everywhere

Distinctive Capabilities

integrated business processes and capabilities that together serve customers in different ways from competitiors

R²

coefficient of determination, measures how much of the ups and downs (variation) in one thing (dependent) can be explained or predicted by another thing (independent) (ranges from 0 to 1)

nominal variable

label, category, not having a numerical value (favorite color or type of pet you have)

data partitioning

dividing data into subsets of training and validation

histogram

graph that shows how often numbers fall into specific groups or ranges (0-10, 11-20, etc.), looks like spectrum analyzer soundwave visual almost

mosaic

a graphical visualization of a contingency table that displays the relationship between two or more categorical or ordinal variables

box plot

a simple way to summarize a long list of numbers using just five key points