Business Analytics and Data Visualization: Key Concepts and Techniques

Business analytics

The scientific process of transforming data into insight for making better decisions

Descriptive analytics

Describes what has happened in the past using statistics and data visualization

Predictive analytics

Uses models built from past data to predict the future or measure the impact of one variable on another

Prescriptive analytics

Finds the best course of action to take using optimization models

Crosstabulation

A type of table used to describe data involving two variables simultaneously

Scatter chart

Shows the relationship between two quantitative variables; can include a trendline

Trendline

A line added to a scatter chart that approximates the relationship between two variables

Line chart

Displays quantitative data collected over time with points connected by a line

Bar chart

Uses horizontal bars to compare categorical data

Column chart

Uses vertical bars to compare categorical data

Bubble chart

Visualizes three quantitative variables in 2D using x-axis, y-axis, and bubble size as the third variable

Heat map

A 2D representation using shades of color to indicate magnitude; built in Excel using Conditional Formatting

Stacked column chart

Compares relative values of multiple quantitative variables within the same category stacked on top of each other

Clustered column chart

Alternative to stacked column chart that places bars side by side instead of stacking

Time series

A sequence of observations on a variable measured at successive points in time

Horizontal pattern

Data fluctuates randomly around a constant mean with no trend or seasonality

Trend pattern

A gradual upward or downward shift in values over a longer period of time

Seasonal pattern

A recurring pattern that repeats over successive periods of time

Trend and seasonal pattern

A time series that exhibits both a gradual shift over time and a repeating recurring pattern

Dummy variable

A 0 or 1 variable used to represent a categorical variable in regression; k categories require k−1 dummy variables

Reference category

The omitted category in dummy variable regression whose effect is captured by b₀

Dependent variable

The variable being predicted or forecasted in regression; denoted by y

Independent variable

The variable used to predict the dependent variable in regression; denoted by x

Linear trendline regression

ŷₜ = b₀ + b₁t where t is the time period; used when a time series has a trend pattern

Qualitative forecasting

Uses expert judgment to develop forecasts; does not require historical data

Quantitative forecasting assumption

The past is prologue — historical patterns are assumed to continue into the future