Scribd
Upload
26 views16 pages

Data Science Presentation SSJ.01

The document presents an analysis of Mahindra & Mahindra's sales and stock prices from 2015 to 2023, focusing on trends, correlation, and regression analysis. It highlights the data science pipeline used for data cleaning, exploratory data analysis, and regression modeling, revealing a moderate positive correlation between car sales and share prices. Key insights suggest that the regression model can predict share price movements, with recommendations for further refinement and exploration of additional variables.

Uploaded by

laukik.kharote24
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
26 views16 pages

Data Science Presentation SSJ.01

The document presents an analysis of Mahindra & Mahindra's sales and stock prices from 2015 to 2023, focusing on trends, correlation, and regression analysis. It highlights the data science pipeline used for data cleaning, exploratory data analysis, and regression modeling, revealing a moderate positive correlation between car sales and share prices. Key insights suggest that the regression model can predict share price movements, with recommendations for further refinement and exploration of additional variables.

Uploaded by

laukik.kharote24
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

Analysis of Financial Data: Stock Prices and Sales

"Exploring Trends, Correlation, and Regression Analysis."

Presented by:

Name G.R. No. Roll No


Relekar Ronit Swapnil 12420137 31
Kharote Laukik Gopal 12420138 32
Joshi Amey Dhananjay 12420170 35
Jadhav Suraj Santosh 12420222 42
Shakil Abdul Hamid Shaikh 12420274 48

Under Guidance of
Prof. Dr. Jayant. V. Kulkarni
MAHINDRA & MAHINDRA (M&M) OVERVIEW

1)Established: 1945, Ludhiana, India.0

2)Founders: J.C. Mahindra, K.C. Mahindra, Malik Ghulam Muhammad.

3)Core Business Areas- 1) Automotive

2) Agriculture

3) Aerospace & Defence.

4) Renewable Energy

4)Global Presence -Operates in 100+ countries across 5 continents. Market Leader:


SUVs and tractors in India. Recognized on Forbes Global 2000 for innovation and
sustainability.
OBJECTIVES OF ANALYSIS
 OBJECTIVES :-
• Analyse the trends in Mahindra's sales and stock prices.
• Understand the relationship between sales and share price.
• Use corelation and regression analysis to derive insights.

 KEY POINTS :-
1. Data Cleaning
2. Data Science Pipeline
3. Descriptive Statistics and Distributions
4. Hypothesis Testing
5. Regression Analysis
6. Evaluation Metrics
7. Visualization Tools
Data Science Pipeline

• ‘The process and tools used to gather raw data from multiple sources, analyze it, and present the results in an
understandable format.’

• Steps in Data Science Pipeline


i. Data Collection – Gather data from sources like databases, APIs, or sensors.
ii. Data Cleaning – Fix errors (as shown above).
iii. Exploratory Data Analysis (EDA) – Understand data patterns.
iv. Feature Engineering – Create or modify features to improve analysis.
v. Modeling – Use algorithms like regression or classification.
vi. Evaluation – Test how well the model performs.
vii. Deployment – Use the model in real-world applications.
PANDAS: DATA ANALYSIS TOOL
 What is Pandas?

• Pandas is a powerful Python library for data analysis and manipulation.

 Key Features:

• Handles structured data efficiently using Data Frames.


• Provides tools for data cleaning, transformation, and aggregation.
• Supports reading and writing data in various formats (CSV, Excel, SQL, etc.).

 Common Uses:

• Data preprocessing and cleaning.


• Exploratory data analysis (EDA).
• Time-series analysis and statistical operations.
NUMPY: NUMERICAL COMPUTING LIBRARY
What is NumPy?
• NumPy (Numerical Python) is a powerful library for numerical computing in Python.

Key Features:
• Provides multi-dimensional arrays (ndarray) for efficient numerical operations.
• Supports vectorized computations for high performance.
• Includes mathematical functions like linear algebra, statistics, and Fourier transforms.

Common Uses:
• Handling large datasets efficiently.
• Performing mathematical and statistical operations.
• Serving as a foundation for libraries like Pandas, SciPy, and TensorFlow.
DATA OVERVIEW :- SALE’S AND SHARE PRICE
 SALE’S DATA:-
1) Timeframe:- Mentioned the years covered (2015-2023)
2) Frequency :- Quarterly data.

 SHARE PRICE DATA:-


1) Timeframe:-2015-2023(same as a sales data).
2) Metrics :- Open, Close, High, Low, and Mean price.
Data Overview

Fig. 1 : Quarterly Sales Data Fig. 2 : Quarterly Stock Price


DATA CLEANING PROCESS
Data cleaning is the process of identifying and fixing errors or inconsistencies in a dataset to make it usable for analysis. Raw data
often contains missing values, duplicates, or irrelevant information that can affect the results.

 STEPS TAKEN :-
a. Handling missing values :-
Checked and filled missing values.
b. Standardizing Formats :-
Converted columns like "Date" to proper datetime format.
c. Derived Metrics :-
Added a new column for Average Stock Price using,

 CHALLENGES :-
a. Any inconsistencies in data (e.g., unexpected missing quarters or outliers).
b. Limited timeframe of data.
EXPLORATORY DATA ANALYSIS

 A line chart showing Car Sales Trend Over Time (Mahindra).


From this graph we can discuss sales fluctuations during different
quarters and years. The chart highlights trends, including declines
and recoveries.

 A graph also displays the co-relation between sales and share price.

 Example:-"As car sales increase, share prices tend to rise.“


REGRESSION ANALYSIS
 Code Overview:

1. Implemented regression model using sklearn’s Linear Regression.

 What the plot tells us

1. Correlation:

1. If the blue points are close to the red line, it indicates a strong linear
relationship between car sales and share price.

2. If the points are widely scattered, the relationship might be weak.

2. Trend:

1. The positive slope of the red line suggests that as car sales increase, the share
price also tends to increase.
REGRESSION ANALYSIS
 Regression Equation: Share Price=0.0881×Car Sales−1320.74

 Performance Metrics: Include the calculated metrics:

Example:
•MSE: 123456.78 (adjust with your value)
•R²: 0.85 (adjust with your value)
Coefficient: 0.0881
 Key Takeaways:
Intercept: -1320.74
•The regression model shows a positive relationship between car sales and share price.
•For every additional unit sold, the share price increases by 0.0881.
•The model explains approximately 85% of the variance in share price, suggesting a strong fit.

 For example, if car sales increase by 10,000 units, the share price is expected to
increase by: ΔShare Price=0.0881×10000=881
INSIGHTS
 Key Takeaways:

• Sales show a moderate positive correlation with share prices.


• The regression model provides useful insights into predicting price movements.
• Some unexpected results might indicate external factors influencing stock performance.

 Next Steps:

• Further refine the model by including more variables.


• Conduct additional statistical tests to validate findings.
• Explore different regression techniques for improved accuracy.
CONCLUSION

• The analysis identified a moderate positive correlation between Sales and Close prices.
• The regression model effectively predicts share price trends with reasonable accuracy.

 Next Steps:

• Investigate additional variables for enhanced predictions.


• Implement advanced regression techniques to reduce errors.
• Explore real-world factors impacting the observed trends.
CHALLENGES AND FUTURE SCOPE
• Lack of data i.e. ( If monthly/weekly data present ) gives more accuracy

• Other events also affect the stock price

• Analysis can be better using multi point regression


REFERENCES

• Websites like screener, tijori, money control

• Trading view charts

• Company financial results from official web

You might also like