U.S. Employment and Training Administration
Initial Unemployment Claims
Number, Seasonally Adjusted
Weekly, Ending Saturday
4-week moving average of initial claims measures the average number of unemployment claims filed per week in the US for the past 4 weeks. This is indicative of overall economic health, availability of jobs, and economic resessions/depressions, and can provide a more stable value than the per week datasets as it smooths out short term fluctuations.
The data shows autocorrection and a non-normal distribution. The data should be differenced. While the Order Norm transformation, provides the best normality, the Boxcox variable will also perform well.
Data is able to be distributed by time but not by geography. The roll up method used is Sum.
4-Week Moving Avg of Initial Unemployment Claims
Auto Correction Function
Auto Correlation Function After Differencing
Partial Auto Correlation Function
Seasonal and Trend Decompostion
Data shows autocorrectation indicating a need for differencing
The ACF indicates 1 order differencing is appropriate.
Further differencing is reccommended
The Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test, KPSS Trend = 0.94 p-value = 0.01 indicates that the data is not stationary.
The Shapiro-Wilk test returned W = 0.92 with a p-value =0.00 indicating the data does not follow a normal distribution.
A skewness score of 0.47 indicates the data are fairly symmetrical.
Hartigan's dip test score of 0.03 with a p-value of 0.03 inidcates the data is multimodal
Statistics (Pearson P/ df, lower => more normal)
U.S. Employment and Training Administration, 4-Week Moving Average of Initial Claims [IC4WSA], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/IC4WSA, December 16, 2019.