Business Modelling
Unit 1
Data Classification
PDF
Data classification is the process of organizing data into categories or groups based on
shared characteristics, such as type, source, presentation, or content, to simplify analysis
and interpretation.
It helps in identifying patterns, making decisions, and applying appropriate statistical tools.
1. On the Basis of Nature of Variable
This type of classification organizes data based on its form and measurement style:
Quantitative Data:
Data that represents numerical values, which can be measured or counted. It answers
questions like "how much" or "how many.
Example: A person's height (5.9 feet), the number of cars in a parking lot (20).
Qualitative Data:
Data that describes characteristics or qualities. It cannot be measured in numbers but can
be categorized.
Example: Eye color (blue, green), type of music (classical, jazz)
Discrete Data:
Data that consists of distinct, separate values, usually whole numbers, with no fractions.
Example: Number of students in a class (30), number of cars in a garage (5).
1
Continuous Data:
Data that can take any value within a range, including fractions and decimals.
Example: The temperature of a room (22.5°C), weight of a person (60.8 kg).
Chronological or Temporal Data:
Data organized according to time. It is useful for identifying trends and changes over
periods.
Example: Monthly sales records, annual rainfall data.
Geographical or Spatial Data:
Data categorized by geographic locations or regions. It is used in mapping and regional
analysis.
Example: Population distribution by country, rainfall across different states.
2. On the Basis of Source of Collection
This classification depends on how and where the data is obtained:
Primary Data:
Data collected first-hand by the researcher for a specific purpose. It is original and often
gathered through experiments, surveys, or direct observations.
Example: Data collected through a customer satisfaction survey conducted by a company.
Secondary Data:
Data that has already been collected and published by someone else. It is reused by
researchers for their own studies
Example: Data from government reports, newspapers, or online sources.
3. On the Basis of Presentation
This classification focuses on how the data is organized and displayed.
2
Grouped Data:
Data organized into groups or intervals to simplify analysis. This is often used for large
datasets.
Example: Income distribution grouped as ₹10,000–₹20,000, ₹20,001–₹30,000, etc.
Ungrouped Data:
Raw data shown in its original form without any grouping. It is more detailed but harder to
analyze.
Example: Scores of 10 students: 85, 90, 88, 92, 75, 80, 70, 95, 60, 89.
4. On the Basis of Content
This classification is based on the number of characteristics used to divide the data:
Simple Classification:
Divides data into two distinct groups based on one characteristic.
Example: Gender (Male/Female), Residential area (Urban/Rural).
Manifold Classification:
Divides data into multiple categories based on several characteristics.
Example: Classifying students by both age group (10–15 years, 16–20 years) and gender
(Male/Female).
Steps in a Statistical Study
Statistical Study:
3
A statistical study involves the collection, analysis, interpretation, and presentation
of data to draw meaningful conclusions. It helps make decisions based on evidence
and provides insights into patterns and relationships within data.
1. Define the Goal of the Study Precisely
The first step in any statistical study is to clearly define the objective or purpose.
This ensures that the entire process is focused on answering a specific research
question or hypothesis.
Without a well-defined goal, the remaining steps may become irrelevant or
misdirected.
2. Choose a Representative Sample from the Population
After defining the objective, selecting a representative sample is crucial.
A sample that accurately reflects the characteristics of the population allows for
valid inferences to be made.
The sampling method (e.g., random sampling, stratified sampling) is determined to
ensure unbiased representation.
3. Collect Raw Data and Summarize
Data collection involves gathering information from the sample.
This data is then summarized through methods such as descriptive statistics,
organizing it in a meaningful way.
Raw data might be summarized using measures like mean, median, and mode,
which provide initial insights.
4. Use the Sample Statistics to Infer Population Parameters
Once data is collected and summarized, statistical analysis is conducted.
4
This involves applying statistical methods to estimate population parameters based
on sample statistics.
Techniques such as hypothesis testing, confidence intervals, and regression
analysis are commonly used.
5. Draw Conclusions
Finally, the results are interpreted and conclusions are drawn.
These conclusions are based on the analysis of data, where insights are
communicated in a clear and meaningful manner.
The findings are reported, often accompanied by recommendations or decisions
derived from the study.
5
Frequency distribution
Pdf and material
Definition:
Frequency distribution is a statistical method used to organize raw data into a table that
displays the frequency (or count) of occurrences of each unique value or group of values in
a dataset. It helps in understanding the distribution, patterns, and trends within the data.
Key Terms in Frequency Distribution
1. Frequency (f): The number of times a data point appears in the dataset.
2. Class Interval: A range of values used to group the data, particularly in grouped
frequency distribution.
3. Lower and Upper Limits: The smallest and largest values within a class interval.
4. Class Boundaries: Actual limits between adjacent intervals to avoid gaps in data
representation.
5. Midpoint (Class Mark): The average of the lower and upper limits of a class
interval.
Midpoint=Lower Limit+Upper Limit/ 2
6. Cumulative Frequency: The running total of frequencies, either starting from the
smallest (less than cumulative frequency) or the largest (greater than cumulative
frequency).
7. Relative Frequency: The proportion of the frequency of a class relative to the total
frequency. Relative Frequency=Frequency of a Class / Total Frequency
Steps to Construct a Frequency Distribution Table
1. Organize Data:
○ Collect raw data and sort it in ascending order.
2. Determine Range:
○ Calculate the difference between the maximum and minimum values.
6
Range=Maximum Value−Minimum Value
3. Decide Number of Classes (k):
○ Use the Sturges' Rule for determining the number of classes:
k=1+3.322logn
where n is the total number of observations.
4. Calculate Class Width:
○ Divide the range by the number of classes and round up.
Class Width=Range / k
5. Create Class Intervals:
○ Start with the lowest value and form intervals of equal width.
6. Count Frequencies:
○ Tally the number of data points falling within each class interval.
Importance of Frequency Distribution
1. Simplifies Data: Converts raw data into a structured and interpretable format.
2. Highlights Patterns: Helps identify trends, peaks, and variations in data.
3. Aids Comparison: Facilitates comparing different datasets or groups.
4. Supports Analysis: Essential for calculating measures like mean, median, and
mode.
Graphical representation
Pdf
Analyzing Graphs in Statistics
Definition:
Graphical analysis is the process of examining and interpreting visual
representations of data to identify trends, relationships, outliers, and key
7
insights. It is a cornerstone of statistical study, enabling quick comprehension
of data patterns and distribution.
Steps to Analyze Graphs
1. Identify the Graph Type
○ Recognize the type of graph: bar chart, histogram, pie chart, scatter plot, etc.
○ Determine the data type (categorical or numerical) and the graph's purpose:
■ Comparison
■ Trend analysis
■ Distribution
■ Correlation or proportion analysis
2. Understand the Components
○ Title and Labels: Check for clarity in the title and axis labels.
○ Legend: Interpret colors, markers, or line styles.
○ Axes: Analyze the scales (linear, logarithmic), units, and ranges.
○ Data Points/Bars/Lines: Observe how the data is represented visually.
3. Observe Patterns
○ Trends: Check for increasing, decreasing, or cyclical trends.
○ Peaks and Troughs: Identify maximum and minimum values.
○ Clusters: Look for grouping of data points that may indicate relationships.
○ Uniformity or Gaps: Check for consistency or missing data.
4. Evaluate Outliers and Anomalies
○ Highlight data points that deviate significantly from the main pattern.
5. Summarize Insights
8
○ Conclude by summarizing the relationships, patterns, or distributions
observed.
Key Patterns to Identify
1. Trends
○ Steady increase, decrease, or cyclical changes.
2. Symmetry
○ Balanced distributions (e.g., normal distribution curves).
3. Skewness
○ Right Skew: More data points are concentrated on the left, with a long tail
on the right.
○ Left Skew: Opposite of right skew.
4. Outliers
○ Isolated points far removed from the main distribution.
5. Correlation
○ Positive, negative, or no relationship between variables.
Importance of Analyzing Graphs
1. Simplifies Complex Data
○ Converts large datasets into comprehensible visual formats.
2. Reveals Patterns and Trends
○ Helps detect hidden relationships or recurring cycles.
9
3. Supports Decision-Making
○ Provides evidence for strategies and predictions.
4. Identifies Anomalies
○ Outliers and inconsistencies can be flagged for deeper analysis.
5. Enhances Communication
○ Aids in presenting findings clearly in reports or presentations.
Example of Graph Analysis
Dataset: Monthly Sales in USD
Month Sale
January 5000
Februar 7000
y
March 6500
April 8000
May 9000
Line Graph Analysis:
1. Trend: Sales consistently increase, with a slight dip in March.
2. Peak: The highest sales occur in May ($9000).
3. Insight: Sales growth is steady, indicating a potential increase in market demand.
10
Measures of Center
Link
Definition:
Measures of center, also known as measures of central tendency, are statistical tools used
to identify a single value that best represents the center or typical value of a dataset. These
measures simplify complex datasets, making it easier to compare, analyze, and interpret
data trends.
Types of Measures of Center
1. Mean (Arithmetic Average)
● Definition: The mean is the sum of all data points divided by the total number of
data points, providing a balanced representation of the dataset.
● Formula:
Mean(μ)=Σxn\text{Mean} (\mu) = \frac{\Sigma x}{n}
Where:
Σx\Sigma x: Sum of all data points
nn: Total number of data points
● Advantages:
1. Easy to calculate and widely understood.
2. Utilizes all data points, providing a comprehensive measure.
3. Suitable for continuous and discrete numerical data.
● Disadvantages:
1. Highly sensitive to outliers (extreme values can skew the mean).
2. Not suitable for datasets with a skewed distribution.
11
3. May not reflect the dataset accurately when variability is high.
4. Requires numerical data.
● Graphical Representation:
In a normal distribution curve, the mean is located at the center, aligning with the
median and mode.
2. Median
● Definition: The median is the middle value of a dataset when arranged in ascending
order, dividing the dataset into two equal halves.
● Steps to Calculate:
1. For an odd number of observations: Select the middle value.
2. For an even number of observations: Calculate the average of the two middle
values.
● Advantages:
1. Not affected by outliers, making it robust for skewed data.
2. Simple to compute for ordered datasets.
3. Represents the "typical" value in ranked data.
● Disadvantages:
1. Ignores the magnitude of values in the dataset.
2. Cannot be used for categorical data.
3. May not be representative in small datasets.
4. Requires data sorting, which can be time-consuming.
● Graphical Representation:
In a box plot, the median is shown as the line dividing the box into two parts,
12
highlighting the dataset's center.
3. Mode
● Definition: The mode is the value or category that appears most frequently in the
dataset, representing the most common outcome.
● Types:
1. Unimodal: One mode.
2. Bimodal: Two modes.
3. Multimodal: More than two modes.
● Advantages:
1. Can be used for both numerical and categorical data.
2. Highlights the most frequent value or category.
3. Useful for identifying trends in repeated data.
4. Not influenced by outliers.
● Disadvantages:
1. May not exist or may not be unique.
2. Less informative for continuous data.
3. Not suitable for summarizing datasets with evenly distributed values.
4. Provides limited insight in multimodal distributions.
● Graphical Representation:
In a bar graph, the mode corresponds to the highest bar, representing the most
frequent value.
4. Midrange
13
● Definition: The midrange is the average of the smallest and largest values in the
dataset, offering a quick estimate of the center.
● Formula:
Midrange=Min Value+Max Value2\text{Midrange} = \frac{\text{Min Value} +
\text{Max Value}}{2}
● Advantages:
1. Simple and quick to calculate.
2. Useful for estimating the range midpoint.
3. Provides a rough idea of data spread.
● Disadvantages:
1. Highly sensitive to outliers, as it depends only on extreme values.
2. Does not use all data points.
3. Not representative for skewed datasets.
4. Rarely used in statistical analysis due to lack of robustness.
Comparison of Measures
Measur Advantages Disadvantages
e
Mean Comprehensive, widely understood Sensitive to outliers
Median Robust to outliers, suitable for skewed Ignores magnitude of other values
data
Mode Useful for categorical data May not exist or be unique
14
Midrang Simple to calculate Sensitive to outliers, uses only
e extremes
Applications of Measures of Center
1. Mean: Comparing average income, test scores, or production rates.
2. Median: Identifying the central tendency of house prices or salaries in skewed
distributions.
3. Mode: Determining the most popular product or category in marketing.
4. Midrange: Estimating the range midpoint in temperature or sales data.
5. Quartile ;
Measures of Dispersion
link
15
