This project analyzes daily vehicle traffic data from the Baregg Tunnel (November 2003 – November 2005) to determine how accurately daily traffic can be forecasted. The objective is to compare a simple benchmark model (Naïve) against a more structured approach using Linear Regression with trend and weekly seasonality. The data are partitioned into a training period (Nov 2003 – Jun 2005) and a validation period (Jul 2005 – Nov 2005) to evaluate out-of-sample forecast accuracy. Model performance is assessed using standard metrics including ME, RMSE, MAE, MAPE, and MASE. Diagnostic plots are also used to evaluate residual behavior and check for remaining autocorrelation. Results show that incorporating trend and weekly seasonality significantly improves forecast accuracy compared to the Naïve benchmark. The Linear Regression model better captures systematic traffic patterns and produces more reliable daily forecasts. This analysis demonstrates how time series modeling techniques can be applied to real-world transportation data to support forecasting and operational planning decisions.

Decision tree with estimated costs and probabilities if META were to choose or not to go with Nuclear power to meet its speculated 4GWt power consumption by 2030

Dataset: Sankey_data.csv Objective: Visualize Company Layoff's across Different departments Companies: Amazon,Twitter, Meta, Microsoft

Dataset Used: MakeupDB.xlsx, Region.xlsx Visuals: Area Plot, Bubble Chart Focus: 1 - Area Plot for Makeup accessory sales by month over 4 years 2 - Bubble chart for Sales of products by region given their market share

Data Set(s): McDonalds.xlsx - Income v/s Average Visits Dow-1.xlsx - Month for multiple years v/s Returns

Pie Chart Representation Variables: Workforce in different Departments, Salary Ranges DataSet: HR_comma_sep

Data Set: HR_comma_sep Visualizing Lastt_evaluation and Satisfaction_level against employees that LEFT or STAYED at the company

Data Set: HR_comma_sep Rate of Satisfaction at an Eval v/s continued employment