Skip to content

Statistical Modelling and Data Visualization of a Climate Change Dataset (January 1984 to December 2008 ) Sourced from Kaggle

Notifications You must be signed in to change notification settings

maxxhvo/Modelling_Climate_Change_Via_Indicators

Repository files navigation

Modelling-Climate-Change-Via-Indicators:

Exploration of Relevant Parameters of Climate Change Using Multiple Linear Regression and Cross-Validation: A 25 year Study (1984-2008)

One of humanity's most pressing issues today is climate change, characterized by long-term shifts in global temperature and weather patterns. Its consequences include higher global temperatures, more frequent severe storms, and widespread poverty due to displacement, among others. Understanding the root causes of these shifts is crucial for mitigating climate change and reducing its impacts. Crucially this report asks: What are the common elements that affect temperature and weather patterns, and how do they influence global temperatures?

Informed by exploratory analysis, we created a model that utilized the variables ‘Year’, ‘MEI’, ‘CO2’, ‘CH4’, ‘N2O’, ‘CFC.11’, ‘CFC.12’, ‘Aerosols’ by determining a model that best fit Climate Data from 1984-2008. After testing various models, we utilized multiple linear regression which is a supervised statistical method. After conductive statistical analysis and testing, we identified that CH4 and N2O may not have a significant impact on the variable.

DataSet Sourced from: https://www.kaggle.com/datasets/econdata/climate-change/data