Overview

This data analysis portfolio piece will present a preliminary data investigation, showing how a typical data exploration starts from initial data collection to identifying preliminary trends and discussing them. In this analysis, we pull information from U.S. Bureau of Labor Statistics (BLS) databases that includes economic statistics on inflation, prices, unemployment, and pay & benefits. Specifically, in this piece, we explore the change in the Consumer Price Index (CPI) over the years and initialize an investigation into interesting trends that occur around the COVID-19 pandemic.

Data Sources

Our data source is the U.S. Inflation and Unemployment dataset from the U.S. Bureau of Labor Statistics.

These data were accessed via BigQuery public datasets using R-Studio.

Consumer Price Index

“The Consumer Price Index (CPI) is a measure of the average change over time in the prices paid by urban consumers for a market basket of consumer goods and services.” (From the BLS).

CPI Over The Years

Below, the diagram presents the monthly Chained CPI for All Urban Consumers (C-CPI-U) value over time, from December 1999 to December 2021. “The Chained Consumer Price Index for All Urban Consumers (C-CPI-U) is a monthly measure of the average change over time in the prices paid by urban consumers for a market basket of consumer goods and services. The C-CPI-U employs a formula that reflects the effect of substitution that consumers make across item categories in response to changes in relative prices. Index data are available for the U.S. City Average (or national average) only. Indexes are available for the all items index and for selected major subcomponents and aggregates” (From the BLS). Thus, the C-CPI-U avoids overestimating the influence of inflation on consumers compared to the CPI-U in general.

Upon initial investigation, we notice a possible significant change in the C-CPI-U trend at around May, 2020 (Indicated by the red vertical dotted line in the diagram) brought about by the COVID-19 pandemic. Prior to May 2020, we see periodic trends in the C-CPI-U where it is above or below the average C-CPI-U value, but the overall rate of change of the C-CPI-U was relatively stable, on average. After May 2020, however, we can eyeball a seemingly significant increase to this previously stable C-CPI-U rate of change. We will save a time series investigation for a later portfolio piece, in which we will attempt to decompose this observed C-CPI-U curve into its trend, seasonal, and random components. For our preliminary analysis, we will focus on investigating the the average trends.

To investigate the effect of COVID-19 on the rate of change of the C-CPI-U, we fit a linear regression with a spline at the identified inflection point of May 2020. This linear fit is represented by a solid blue line on the diagram above. Prior to May 2020, the average rate of change of the C-CPI-U (with standard error) was around 2.1044 (± 0.0025) CPI points per year. After May 2020, this rate of change increased to around 4.6865 (± 0.0023) CPI points per year. This is a 122.7% increase in the rate of change of the C-CPI-U, indicating that after the start of the COVID-19 pandemic in the U.S., the rate of inflation change elevated greatly. Comparing these two slopes via Cohen’s d effect size calculation, we get an effect size of approximately d = 1.5100. In general, a Cohen’s d value greater than 0.8 is considered large; thus, a value of >1 indicates a relatively acute change in how quickly inflation has been rising in the United States around the start of the COVID-19 pandemic.

Did the COVID-19 Pandemic Effect All Products Equally?

The BLS database also provides the same C-CPI-U information broken down by product type. Let’s explore the data, look for interesting trends, and focus on one or two aspects for initial investigation.

Due to the large number of categories and only so many colors available to use, we add a grey background to increase the contrast for our plotted lines. Though this diagram is quite busy, we can get important information at a glance and then focus on some interesting aspects. We see some product categories out of the 27 total experienced price deflation, indicated by the curve staying below the CPI = 100 line. We also see that the acute effect of COVID-19 seen for “All items” in not seen in all product categories. To get a better look at the possible effects of the COVID-19 pandemic, Let’s focus in on an area centered around May 2020, ranging from October 2018 to December 2021.

This diagram is still quite busy, and it is not easy to pick out each trend, even with the grey background added to provide better color contrast. Thus, let us make individual C-CPI-U diagrams for each Product Category, fitting a linear regression with a spline at the identified inflection point of May 2020 to help evaluate the effect of the COVID-19 pandemic. We utilize loops here to make all 27 plots in short time.

Let us also visualize the effect of the COVID-19 pandemic by extending the linear regression prior to May 2020 represented in the plots below as a dashed blue line and calculating the angle between the linear regression before and the linear regression after the inflection point. This will give us a measure of the degree to which the rate of change of the C-CPI-U was effected by the COVID-19 pandemic within each Product Category. We will call this the “Deflection Angle.”

You can skip down to the Discussion Section to read about the initial impressions of the differential effects of the COVID-19 pandemic on C-CPI-U by clicking here.

C-CPI-U Diagrams Focused around May 2020 for each Product Category

Click a Product Category below to jump to that plot, or scroll down to puruse them all. NOTE: The deflection angle given may not match the angle estimated by eye visually on the presented graphs due to the difficulty of maintaining the same axes scales for each graph.

Education and communication

Household furnishings and operations

Housing

Medical care

Medical care commodities

Medical care services

New vehicles

Nondurables

Other goods and services

Private transportation

Public transportation