class: center, middle, inverse, title-slide .title[ # Predicting Car MPG with a Shiny App ] .author[ ### Jay Jamison ] .date[ ### August 2024 ] --- ### Slide 1: Introduction ## A Simple, Powerful Data Product Welcome! In this presentation, we’ll explore a data product that is both easy to use and highly impactful: an interactive web application designed to predict a car’s fuel efficiency. By combining clean design with robust predictive modeling, this tool transforms complex data into actionable insights—helping users make smarter, data-driven decisions about vehicle performance. --- ### Slide 2: The Problem ## Why predict MPG? Fuel efficiency, measured in Miles Per Gallon (MPG), is one of the most important considerations for car buyers, manufacturers, and policy makers alike. A vehicle’s MPG directly impacts ownership costs, environmental footprint, and long-term value. Our goal is to make fuel efficiency predictions simple, fast, and accessible. By leveraging just one easily obtainable attribute—a car’s weight—we can provide users with a reliable MPG estimate. This streamlined approach removes the complexity of traditional models while still delivering insights that help drivers, dealers, and researchers make smarter, data-driven decisions. --- ### Slide 3: The Data and Method ## My Approach To demonstrate this concept, we turn to the classic mtcars dataset, a well-known resource in the data science community containing specifications and performance metrics of car models from the 1970s. From this dataset, we focus on the relationship between vehicle weight and fuel efficiency (MPG). By applying a simple linear regression model, we can capture the intuitive but powerful insight that heavier cars generally consume more fuel. This approach strikes a balance between simplicity and interpretability: the model is easy to explain, quick to compute, and effective at showing how even a single predictor—car weight—can provide valuable estimates of real-world performance. **Linear Model in Action:** ```r # Load the mtcars dataset data(mtcars) # Fit the linear model fit <- lm(mpg ~ wt, data = mtcars) # Display the model summary summary(fit) ``` --- ### Slide 4: The Shiny App ## My Solution: The Shiny App We’ve transformed this model into a powerful, user-friendly Shiny application that makes predictive analytics accessible to anyone. With just a few clicks, users can explore the relationship between car weight and fuel efficiency in real time. ✨ Key Features: Interactive Input: Adjust a simple slider to set the car’s weight. Instant Insights: The app instantly returns a fuel efficiency (MPG) prediction, powered by our regression model. Accessible to All: No technical background required—just an intuitive interface that delivers clear, actionable results. This Shiny app bridges the gap between data science and everyday decision-making, turning a statistical model into a tool that anyone—car buyers, enthusiasts, or researchers—can use to gain insight and make smarter choices. --- ### Slide 5: Conclusion & Call to Action ## Next Steps What you’ve seen today is just the first step. Our simple weight-based model proves that even one variable can generate valuable insights—but the potential goes much further. Imagine extending this approach to include horsepower, number of cylinders, transmission type, or even modern datasets with hybrid and electric vehicles. Each new factor adds depth, accuracy, and relevance—transforming a simple demo into a full-featured predictive engine for the future of automotive decision-making. This project opens the door to endless possibilities—what we’ve built is just the beginning. Try out the app for yourself! * **App Link:** `https://jhjamison.shinyapps.io/dataProductsFinalProject/` * **GitHub Code:** `https://github.com/LouieJamison/dataProductsFinalProject` Thanks for your time!