getting Significant seasonality on a straight line I'm running a Significant seasonality test on my data with isSeasonal {seastests} function.
I'm not sure why I'm getting a TRUE answer for the following data.
library(ggplot2)
library(seastests)
pop_ts = ts(economics$pop,frequency = 12, start = c(1967,7))
isSeasonal(pop_ts)

[1] TRUE


 A: The data is seasonal by the looks of things.
I just read the data in and plotted the residuals of a linear model and there is definitely some bump every year (seen most clearly between years 30 and 40). I'm guessing there's some very small seasonal effect on top of the predominantly linear trajectory that you're just not seeing by eyeballing the data. Not sure how small a seasonal effect has to be in order to be detected by that command though.

A: after receiving the data from the dropbox , I have some interesting things to report using AUTOBOX , a time series analusis package that I have helped to develop.
To some it would appear that differencing is the remedy for the non-stationarity BUT not too all ...Is a time trend a substitute for first differencing? discusses the need for incorporating deterministic time trends ...as needed in this example.
It definitely has arima seasonal structure ..... and some short term arima structure AND 5 different trend point changes .  AND non-constant error variance requiring weighted least squares  following http://docplayer.net/12080848-Outliers-level-shifts-and-variance-changes-in-time-series.html
My 81 year old eye failed to identify that there were 5 trends      .... In addition there were seasonal parameters and an error variance change breakpoint .  following http://docplayer.net/12080848-Outliers-level-shifts-and-variance-changes-in-time-series.html
The forecast plot is here 
The residual ACF suggesting sufficiency is here 
