All Times EDT
Virtual
Analysis and Forecast of US Workable Adolescent Male Unemployment (310011)
*Wenjing Li, University of California, Santa BarbaraKeywords: times series, unemployment, forecast
Unemployment has always been an important factor to determine economic behavior in a country. This objective of this project is to predict number of unemployment in thousands from 1975 to 1976 using the method of times series. We transferred and differenced the original data to obtain a stationary data. Using the stationary data, we build a proper SARIMA model with the lowest AICc and BIC scores. Further, we con- ducted diagnosis check and spectral analysis for our final model. At the end, we forecasted number of unemployment in thousands from January 1975 to December 1975. The forecasted values are within the 95% confidence interval, and these values are very close to the original data, which many of them overlap with the original data.