Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/sayed-ashfaq/yulu-hypothesistesting

In this project, I analyzed Yulu bikes' usage patterns by performing data visualization and applying various hypothesis testing methods. This included statistical tests, chi-square tests, variance testing, QQ plots, and checking for normal distribution to uncover insights related to customer usage and behavior.
https://github.com/sayed-ashfaq/yulu-hypothesistesting

anova-test chisquare-test hypothesis-testing pandas-dataframe python scipy-stats seaborn ttest

Last synced: 20 days ago
JSON representation

In this project, I analyzed Yulu bikes' usage patterns by performing data visualization and applying various hypothesis testing methods. This included statistical tests, chi-square tests, variance testing, QQ plots, and checking for normal distribution to uncover insights related to customer usage and behavior.

Awesome Lists containing this project

README 

        
# Yulu Shared Electric Cycle Demand Analysis  



## About Yulu  

Yulu is India's leading micro-mobility service provider, offering eco-friendly and convenient vehicles for daily commutes. With a mission to eliminate traffic congestion, Yulu provides shared, solo, and sustainable commuting solutions through a user-friendly mobile app.  



Yulu zones are strategically located near metro stations, bus stands, residential areas, office spaces, and other key locations to make first and last-mile connectivity smooth, affordable, and convenient.  



Recently, Yulu has faced significant revenue dips and has engaged a consulting team to identify the factors driving the demand for their shared electric cycles in the Indian market.  



---



## Objective  

The primary goals of this analysis are to:  

1. Identify significant variables that influence the demand for shared electric cycles.  

2. Evaluate how well these variables explain the overall demand.  



---



## Dataset  

The dataset contains data on daily electric cycle rentals and related environmental and seasonal factors.  



**Key Features**:  

- **`datetime`**: Date and time of observation.  

- **`season`**: Season of the year (1: Spring, 2: Summer, 3: Fall, 4: Winter).  

- **`holiday`**: Whether the day is a holiday (0 = No, 1 = Yes).  

- **`workingday`**: Whether the day is a working day (0 = No, 1 = Yes).  

- **`weather`**: Weather conditions categorized as:  

  - 1: Clear, Few Clouds, Partly Cloudy  

  - 2: Mist, Cloudy, or Few Clouds  

  - 3: Light Snow or Rain  

  - 4: Heavy Rain or Snow  

- **`temp`**: Temperature in Celsius.  

- **`atemp`**: Perceived temperature in Celsius.  

- **`humidity`**: Humidity percentage.  

- **`windspeed`**: Wind speed.  

- **`casual`**: Count of casual (non-registered) users.  

- **`registered`**: Count of registered users.  

- **`count`**: Total rental bike count, including both casual and registered users.  



---



## Process Overview  



### 1. **Exploratory Data Analysis (EDA)**:  

- Visualized relationships between demand (`count`) and features like `season`, `weather`, `temp`, and `windspeed`.  

- Performed Bi-Variate Analysis to uncover trends and correlations.  



### 2. **Statistical Testing**:  

- **2-Sample t-Test**: Compared means of bike rentals across different groups (e.g., holidays vs. working days).  

- **ANOVA**: Assessed differences in bike demand across seasons.  

- **Chi-Square Test**: Evaluated relationships between categorical variables such as `weather` and `season`.  



### 3. **Key Observations**:  

- Seasonal variations significantly impact demand, with higher rentals observed in fall and summer.  

- Weather conditions have a noticeable effect, with clear weather driving higher demand.  

- Working days tend to see more registered users, while holidays attract more casual users.  



---



## Key Insights  



1. **Seasonality in Demand**:  

   - Rentals peak during favorable weather conditions (fall and summer seasons).  



2. **Weather Influence**:  

   - Clear weather positively correlates with increased bike rentals, while heavy rain or snow significantly reduces demand.  



3. **User Behavior**:  

   - Registered users contribute consistently higher rentals on working days, while casual users drive demand on holidays.  



---



## Tools and Libraries  

This project utilized the following tools:  

- **Python**:  

  - `Pandas` for data manipulation.  

  - `Matplotlib` and `Seaborn` for data visualization.  

  - `Scipy` and `Statsmodels` for statistical testing.  

- **Jupyter Notebook**: For interactive analysis and documentation.  



---



## Repository Structure  

- **`data/`**: Contains the dataset used for analysis.  

- **`notebooks/`**: Jupyter Notebooks documenting the analysis process.  

- **`visualizations/`**: Saved plots and charts.  

- **`README.md`**: Overview of the project (this file).  



---



## Next Steps  

Future work could include:  

1. **Feature Engineering**: Derive new features (e.g., lag variables or rolling averages) to enhance predictive insights.  

2. **Predictive Modeling**: Use machine learning models to forecast bike demand.  

3. **Deep Dive into User Segments**: Analyze behavioral patterns across casual and registered users for targeted marketing strategies.  



---



## Acknowledgments  

- **Dataset Source**: Provided by Scaler for this analysis.  

- **Python Libraries**: Thanks to the open-source Python community for providing versatile data analysis tools.  



---



## License  

This project is licensed for educational and non-commercial use only. If utilizing any part of this repository, please credit the author.