This is the final project for the University of San Diego’s ADS 506 Time Series course. The technical workbook can be found at the main GitHub repository.
The preliminary objective of this study is to accurately forecast future unemployment rates in California. From which, the secondary objective was to subjectively deduce the likelihood of a recession occurring in the upcoming year (2023), through the forecasted unemployment rate values. Multiple sources have stated that a recession was likely to occur in 2023. Thus, this study aimed to identify whether unemployment rates, alone, could be indicative of an upcoming recession.
- Exploratory Data Analysis (EDA)
- Pre-Processing
- Series Characterization
- Data Partitioning
- Forecasting Methods
- RStudio - Version 1.4.1717
- R - Version 4.1.1
- The California Unemployment Rate dataset is provided by Federal Reserve Economic Data (FRED).
- The following forecasting methods were assessed for optimality in forecasting unemployment rates in California:
- Naive Forecast
- Double-Exponential Smoothing
- Holt-Winters Method
- ARIMA
- ARFIMA
- Clone this repository (For help, refer to this tutorial)
- Raw data is kept in the GitHub repository and Kaggle.
- Data preprocessing, exploratory data analysis, and forecasting methodology are in the R Markdown File.
- Nicholas Lee