BUSINESS ANALYTICS
REVIEW OF BASIC
BUSINESS ANALYTICS
METHODS
Chapter 5
Dr. Esraa Ahmad
TEAM
1. Ahmed Mohamed Mahmoud Ali
[Link] Mahrous Samuel
[Link] Maged Moaed
[Link] Wagih Gad Elsayed
[Link] Ahmed Mosa
[Link] Nagy Alzeer
TOPICS
1. Introduction
2. Exploratory Data Analysis (EDA)
3. Summary Statistics
4. Data Exploration vs. Data Presentation
5. Data Visualization
[Link] Plots
7. Histograms
[Link] Plots
[Link] Analytical Processing (OLAP)
[Link] Relationships Between Variables
[Link] Series Analysis
[Link]
#1 INTRODUCTION
Business Analytics: Refers to technologies and practices for
continuous iterative exploration of past business performance.
Core Objective: Help businesses gain insights and drive planning
by using data.
Key Focus Areas:
• Exploratory Data Analysis (EDA): Gaining an understanding of
the data.
• Statistical Models: Predicting outcomes and identifying trends.
• Data Visualization: Translating data into graphs and charts for
clearer understanding.
#2 EXPLORATORY DATA ANALYSIS (EDA)
EDA is the process of using statistical graphics and summary
statistics to identify key patterns, trends, or anomalies in data.
Techniques: Importance:
• Summary Statistics: Numerical • Helps uncover structure in data.
descriptions of key data properties. • Identifies important variables.
• Graphical Methods: Visualizations • Detects outliers and anomalies.
like histograms and scatter plots. • Assists in selecting the appropriate
• Correlation Analysis: Understanding statistical tools for modeling.
the relationships between variables.
#3 SUMMARY STATISTICS
Purpose: Provides a quantitative summary of data properties.
Key Metrics:
• Mean: The average of all data points.
• Median: The middle value when data is ordered.
• Mode: The most frequently occurring value.
• Variance & Standard Deviation: Measures of the spread of data
around the mean.
• Range: Difference between the maximum and minimum values.
Examples:
• In sales data, mean sales give the average sale amount, while
standard deviation shows how spread out individual sales are.
#4 DATA EXPLORATION
VS. DATA PRESENTATION
Data Exploration:
• Purpose: Focuses on uncovering insights from raw data for analysis.
• Audience: Data scientists and analysts.
• Techniques: Statistical measures, visualizations (scatter plots, histograms, box
plots), and finding patterns or anomalies.
• Goal: To understand the data deeply, detect outliers, identify correlations, and
choose appropriate models.
#4 DATA EXPLORATION VS.
DATA PRESENTATION
Data Presentation:
• Purpose: Communicates the findings to a non-technical audience (e.g.,
stakeholders, decision-makers).
• Audience: Business managers, executives, and external stakeholders.
• Techniques: Simplified visuals, high-level summaries, and actionable insights.
• Goal: To present the most relevant data in an easily understandable way,
focusing on key points that aid decision-making.
Key Difference:
• Exploration is about digging deep into the data for analysis, while presentation
is about simplifying the findings for decision-making.
#5 DATA VISUALIZATION
Importance:
Visualizing data helps make complex datasets easier to
understand, identify patterns, and present insights effectively.
Common Techniques:
• Scatter Plots: To visualize relationships between two
quantitative variables.
• Histograms: To visualize the frequency distribution of a single
variable.
• Box Plots: To summarize data distribution, showing the median,
quartiles, and outliers.
• Bar Charts & Pie Charts: For categorical data visualization.
#5 DATA VISUALIZATION
Why Use Visualization?
• Humans are better at understanding data visually than through
tables or raw numbers.
• Detects trends, patterns, and outliers that may not be apparent
through summary statistics.
#6 SCATTER PLOTS
A scatter plot: is a graph that uses dots to represent the values of
two different variables, To determine the correlation between
two variables.
Example:
• X-Axis: Represents one variable (e.g., petal width).
• Y-Axis: Represents the other variable (e.g., petal length).
• The distribution of points indicates how closely the variables are related.
Usage:
Scatter plots can be enhanced with additional dimensions by using point color,
size, or shape.
#6 SCATTER PLOTS
#7 HISTOGRAMS
A histogram: is a bar graph-like representation that organizes a group of
data points into user-specified ranges (bins), To show the frequency
distribution of a variable.
Example:
• A histogram of purchase amounts shows how often certain ranges
(e.g., $0-$100) occur.
Key Insights from Histograms:
• Skewness: Histograms can reveal whether data is skewed to the left
(negative skew) or right (positive skew).
• Modality: Can indicate whether data is unimodal, bimodal, etc.
• Outliers: Visualizing extreme values that occur infrequently.
#7 HISTOGRAMS
#8 BOX PLOTS
Box plots (or whisker plots) display data distribution through
their quartiles, highlighting the median and potential
outliers.
Key Elements:
• Median (50th percentile): Middle value of the dataset.
• Quartiles (25th and 75th percentiles): Data split into four
equal parts.
• Whiskers: Indicate the range within the bulk of the data.
• Outliers: Data points that are much lower or higher than
the rest.
#8 BOX PLOTS
Example:
Comparing multiple variables like petal length and width for
different species.
Usage:
Box plots are useful for comparing the distribution of
multiple groups or variables side by side.
#8 BOX PLOTS
#9 ONLINE ANALYTICAL PROCESSING (OLAP)
OLAP is a data analysis method that uses multidimensional data
cubes for rapid analysis.
Key Features:
• Multidimensional Analysis: Data can be analyzed from multiple
perspectives, such as time, geography, or product category.
• Slicing and Dicing: Extracting subsets of data for specific
analysis.
• Roll-Up and Drill-Down: Aggregating data or breaking it down to
different levels of detail (e.g., from years to months).
#9 ONLINE ANALYTICAL PROCESSING (OLAP)
Benefits:
• Speed and flexibility for
decision-making.
• Quickly reveals patterns and
relationships in data.
Example:
Analyzing sales data by time,
region, and product
simultaneously.
#10 ANALYZING RELATIONSHIPS BETWEEN VARIABLES
Types of Analysis:
• Continuous vs. Continuous Variables: Use scatter plots, bin plots, or
hexbin plots to analyze relationships.
• Continuous vs. Categorical Variables: Use jitter plots, box-and-whisker
plots, or bar charts.
Key Questions to Explore:
• Is the relationship linear or non-linear?
• Are there clusters, trends, or anomalies?
Example:
Visualizing household income by region with scatter plots and box-and-
whisker overlays.
#11 TIME SERIES ANALYSIS
Time series analysis involves evaluating data over intervals of time, To detect
trends, cycles, and seasonal variations in the data.
Example:
• Analysis of air travel data from 1949-1960.
• Insights: Shows upward trends, with peaks during specific months
(seasonality).
Insights from Time Series Analysis:
• Identifies long-term trends and cyclical patterns.
• Detects seasonal effects (e.g., peak sales during holidays).
#12 CONCLUSION
Key Takeaways:
• Exploratory Data Analysis (EDA): Vital for understanding data
before applying advanced statistical models.
• Visualization Techniques: Crucial for identifying trends,
patterns, and outliers in the data.
• OLAP & Time Series Analysis: Offer advanced methods for
multi-dimensional and time-based data analysis.
• Business Impact: These methods help in making data-driven
decisions that improve business performance and strategy.
THANK YOU
Any Questions?