Project Name

MLR-On-SharedBikesDemand (Linear Regression Assignment)

Problem Statement:

Multiple linear regression model for the prediction of demand for shared bikes.

• Which variables are significant in predicting the demand for shared bikes?
• How well those variables describe the bike demands?

Steps for MLR

• Reading, understanding and visualising the data
• Preparing the data for modelling (split training and testing data, rescaling etc.)
• Train the model
• Residual Analysis
• Predictions and evaluation on test set

Conclusions

• After data processing the continuous variables: 'temp', 'atemp', 'hum', 'windspeed'
• After data processing the categorical variables: 'season', 'holiday', 'workingday', 'weathersit', 'yr', 'mnth', 'weekday'
• After model training the final model variables: 'temp', 'windspeed', 'weathersit', 'season', 'holiday', 'yr'
• Final model: cnt = 0.1656 + (temp0.4696) + (windspeed-0.1483) + (weathersit-0.1842) + (season0.1245) + (holiday-0.0952) + (yr0.2361)

Technologies Used

• pandas
• numpy
• matplotlib
• seaborn
• sklearn
• statsmodels

Contact

Created by [@Rameshkatiyar] - feel free to contact me!