For this project I created a twelve month forecast for Total Vehicle Sales in the United States using four different methods. These four techniques are called exponential smoothing, decomposition, ARIMA, and multiple regression. To do so I picked one dependent (Y) variable along with two independent (X) variables and collected 80 monthly observations for each variable. This historical data allowed me to create four different forecasting models which predict future Vehicle Sales with low risk of error. The best model according to the lowest error measures was winter’s exponential smoothing method because it had the lowest MAPE along with the lowest RMSE for the fit period as well as the forecast period.
Table of Contents
Executive Summary
Introduction
Body
Exponential Smoothing
Decomposition
ARIMA
Multiple Regression
Conclusion
Project Goals and Topics
The primary objective of this project is to develop an accurate twelve-month forecast for Total Vehicle Sales in the United States by evaluating and comparing four distinct quantitative forecasting methodologies.
- Application of exponential smoothing, decomposition, ARIMA, and multiple regression models.
- Statistical analysis of economic indicators, including non-farm employment and personal saving rates.
- Evaluation of model accuracy using MAPE (Mean Absolute Percentage Error) and RMSE (Root Mean Square Error).
- Residual diagnostics to ensure randomness and the validity of the forecasting models.
Excerpt from the Book
Introduction
I chose the Y variable to be Total Vehicle Sales in the United States because I have a strong interest in the auto industry and would like to work for a German car maker in the future. The auto industry is very vulnerable to the state of the economy because people tend to postpone high-item purchases like a car when times are tough. Therefore, the variables that cause a change in vehicle sales numbers must be indicators of economic performance. In order to forecast the dependent variable Y (Total Vehicle Sales), I chose two independent variables, X1 and X2 that are closely related to Y. These are going to be Employment non-farm and the Personal Saving Rate. The hypothesis I make for the first X variable is that employment numbers are logically related to vehicle sales because the more people are in the workforce, the more people earn an income which is necessary to make high-item purchases like a personal car. The hypothesis for the second X variable is that the personal saving rate has an inverse linear relationship to vehicle sales because the more people hold on to their disposable income, the less spending occurs which hurts vehicle sales numbers.
Summary of Chapters
Executive Summary: Provides an overview of the four forecasting methods applied to Total Vehicle Sales and identifies the winter’s exponential smoothing method as the most accurate model.
Introduction: Explains the choice of variables (Total Vehicle Sales, Employment non-farm, Personal Saving Rate) and establishes the hypotheses regarding their economic impact on vehicle purchases.
Body: Detailed execution and statistical validation of the four forecasting models, including data transformations, residual analysis, and accuracy comparisons.
Exponential Smoothing: Details the application of winter’s method to capture trend and seasonality, confirming it as the best-performing model based on error metrics.
Decomposition: Examines the seasonal components of the vehicle sales data and attempts to improve accuracy through cyclical factor adjustments.
ARIMA: Focuses on making the data stationary through differencing and identifying the appropriate MA model parameters.
Multiple Regression: Evaluates the linear relationships between the dependent variable and the chosen independent variables while incorporating dummy variables for seasonal effects.
Conclusion: Summarizes the final performance metrics for all models and confirms the superiority of the exponential smoothing approach for this specific dataset.
Keywords
Total Vehicle Sales, Economic Forecasting, Exponential Smoothing, Decomposition, ARIMA, Multiple Regression, MAPE, RMSE, Employment non-farm, Personal Saving Rate, Seasonality, Trend Analysis, Time Series, Residual Analysis, Statistical Modeling
Frequently Asked Questions
What is the core focus of this research project?
The project focuses on creating a reliable twelve-month forecast for Total Vehicle Sales in the U.S. by applying and comparing four different forecasting techniques.
What are the primary thematic areas covered?
The study centers on economic forecasting, specifically analyzing how macro-economic variables like non-farm employment and personal savings impact consumer vehicle purchasing behavior.
What is the specific goal of the forecast?
The goal is to determine the most accurate model by evaluating specific error metrics, such as MAPE and RMSE, for both the fit period and the forecast period.
Which statistical methods are utilized in the work?
The author employs exponential smoothing (specifically winter’s method), decomposition, ARIMA modeling, and multiple regression analysis.
What does the main body of the paper address?
The body section details the step-by-step implementation of each model, providing plots, autocorrelation functions, and statistical tests (like the Ljung-Box test) to validate the models.
Which keywords best characterize this work?
Key terms include Time Series, Statistical Modeling, Seasonality, Residual Analysis, and various accuracy metrics such as MAPE and RMSE.
Why was winter’s exponential smoothing selected as the best model?
It was chosen because it achieved the lowest MAPE and RMSE values during both the fit and forecast periods, successfully capturing the inherent trend and seasonality of the data.
How does the project account for seasonality in the regression model?
The project incorporates dummy variables into the regression equation to adjust for significant seasonal patterns observed in the vehicle sales data.
- Arbeit zitieren
- Alexander Hardt (Autor:in), 2013, Total Vehicle Sales Forecast, München, GRIN Verlag, https://www.grin.com/document/279609
