Evaluation of linear and quadratic trend regression models for NovaBrew Coffee Roasters’ monthly revenue data. Comparing the two models to identify the one that most appropriately captures the long-run growth pattern for forecasting and for informing a major capital investment decision.

Hedonic pricing is a method used to estimate the value of a good by decomposing it into the attributes that make it up. In housing markets, the observed house price reflects both structural characteristics and neighborhood characteristics. This report examines the `hprice2` dataset from the `wooldridge` package to investigate the relationship between housing prices and several neighborhood and property characteristics. Research question: What is the impact of the crime rate on the median house price in a given area, holding other factors constant? The analysis uses multiple linear regression to estimate the effect of crime while controlling for other relevant variables.

This report builds a decision tree model on the Titanic dataset to predict passenger survival based on age, sex, class, and fare. The tree is interpreted to extract simple, human‑readable survival rules for different groups (adult males, young males, and female passengers across classes).

This project analyzes daily vehicle traffic data from the Baregg Tunnel (2003–2005) to develop and evaluate forecasting models. A Naïve benchmark model is compared against a Linear Regression model incorporating trend and weekly seasonality. The dataset exhibits strong weekly seasonality and a mild upward trend. Using a validation period (July 2005 – November 2005), model performance is evaluated through RMSE, MAE, MAPE, and MASE. Results show that the Linear Regression model significantly outperforms the Naïve approach by effectively capturing weekly traffic patterns and underlying trend components. Residual diagnostics confirm that model assumptions are satisfied, indicating reliable forecasts. This analysis demonstrates how incorporating seasonality and trend improves forecasting accuracy in real-world transportation data.

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