When People Who Don’t Understand Basic Time Series Analysis Try to Talk Time Series Analysis

From 2019, pay special attention to CoRev’s statistical analysis (at the end):

Reader JBH writes:

“Employment has done marvelously well under this president.”

I laughed and laughed and laughed when I read this. Why? Take a look:

Figure 1: Nonfarm payroll employment (dark blue), and stochastic trend (red). Stochastic trend estimated using 2010-2016 data, and regression of first log difference on a constant. Source: BLS March 2019 employment situation release, and author’s calculations.

We are doing as well as we were from 2010-2016, even after massive tax cuts and the end of spending restraints amounting in the trillions.

But kudos to JBH for livin’ in a fantasy world … it for sure is more pleasant than my world. Thanks, Drumpf!

Update, 4/12/2019, 4:15PM Pacific: Reader CoRev requests:

Can you provide the raw data used?

Here is the dataset used, directly downloaded from FRED.

Update, 4/16/2019, 9AM Pacific: Reader CoRev has provided his check-file on my trend analysis. Without comment, here it is:

[link]

Some three years after CoRev provided his “anomaly analysis”, I still don’t fully understand what he’s doing. It looks like deviations from averages. If anybody can tell me why it validates his view of the world (i.e., my choice of use of stochastic trend and time sample biases against seeing a boom in Trump employment up to that point), please tell me. I am (still) dying to understand.

In the meantime, here is a graph demonstrating the trouble with linear (deterministic) trends as applied to nonfarm payroll employment (reprised from “Why Friends Don’t Let Friends Apply Deterministic Time Trends to Nonfarm Payroll Employment”):

Figure 1: Nonfarm payroll employment, 000’s, s.a. (black), and linear deterministic trends estimated over 20 year subsamples. NBER defined recession dates shaded gray. Source: BLS, May employment situation release, NBER, and author’s calculations.