U.S. Bureau of Economic Analysis
Millions of Dollars, Seasonally Adjusted
Retail sales excluding food services measures disposable income willingness to spend money, and overall economic health.
The data shows autocorrection and a non-normal distribution. The data should be differenced. While the Arcsin transformation, provides the best normality, the Log variable will also perform well.
Data is able to be distributed by time but not by geography. The roll up method used is Sum.
Retail Sales Excluding Food Service - Seas Adj
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.
Following first order differencing, no further differencing is required based on the differenced ACF at lag one of -0.13
The Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test, KPSS Trend = 0.36 p-value = 0.01 indicates that the data is not stationary.
The Shapiro-Wilk test returned W = 0.95 with a p-value =0.00 indicating the data does not follow a normal distribution.
A skewness score of 0.26 indicates the data are fairly symmetrical.
Hartigan's dip test score of 0.04 with a p-value of 0.55 inidcates the data is unimodal
Statistics (Pearson P/ df, lower => more normal)
U.S. Census Bureau, Advance Retail Sales: Retail (Excluding Food Services) [RSXFS], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/RSXFS, December 15, 2019.