zahra-hlioui/Air-Quality-Analysis-R

Conducted an in‑depth study of tropospheric ozone (O₃) concentrations across three Tunisian regions. Using R, I gathered, cleaned, and structured raw environmental datasets, then applied statistical preprocessing, exploratory analysis, and predictive modeling techniques to extract actionable insights on regional air quality trends. # Air-Quality-Analysis-Using-R My project delves into air quality analysis, specifically focusing on ozone tropospheric (O3) concentration data collected from three different regions in Tunisia. Leveraging R, I conducted a series of tasks to preprocess, analyze, and model the data for meaningful insights. Project Tasks: Data Import and Cleaning: Imported project databases and focused on the first three months of 2014. Eliminated unnecessary data for streamlined analysis. Data Preprocessing: Created a new qualitative variable "Mois" based on the Date column. Detected and managed outliers in the dataset. Explored and addressed missing values, proposing a method for imputation. Data Normalization: Scaled variables as needed, justifying choices for normalization. Univariate Analysis: Conducted a comprehensive univariate analysis, summarizing statistics and visualizing data distributions. Bivariate Analysis: Explored correlations between variables using appropriate hypothesis tests, with a focus on the relationship between O3 and other variables. Linear Regression: Performed simple and multiple linear regression analyses to understand the relationship between O3 and other variables. Developed a strategy to enhance the regression model's performance. Reduced dimensionality using principal component analysis. Generalized Additive Models (GAM): Conducted a literature review on GAM models. Implemented GAM models to understand the relationship between O3 and explanatory variables.