A Tale of Two Trends

I find it interesting to see how scientists and pundits are playing two different trends, one of global temperature and the other of US health spending.

The trend in each of these is important.  For global temperature, there are of course models that would indicate global temperature should be increasing with atmospheric CO2 concentrations; with health spending, the underlying theory is less developed, but the implications of changes in health care costs for US government spending and deficits are immense.

For each issue -- global temperature and health spending -- the most recent data observations give rise to speculation on changes in the underlying trend.  For instance, here is a headline on health spending, from this Bloomberg story:

And here is a chart from the Altarum Institute showing the data:

Meanwhile on the global temperature issue, we have all kinds of headlines on whether global warming has "stopped" in the last 15 or so years.  Here is one example of a headline, from the Guardian in the UK:

And here is just one diagram with some relevant global temperature data.  The source for this is the Real Climate blog:  Note the colored lines are different measures of actual temperature, while the black line is a forecast from a certain set of models.

Of course, the key point in all of this is that we have to consider the background variation in the time series in question before we can conclude anything meaningful about a statistically significant change.  The Real Climate chart makes some headway on that front, with confidence intervals around their model forecast.  I won't agree immediately that that method clears up the issue completely, but at least it recognizes the important fact of variability.  As to health care spending, there is also a lot of background variation, and I have seen very little on any statistical inference about changes.

However, I will go out on some ice and make an observation:  The liberal media has been all over the "marked slowdown" in health care costs as if it is for-sure a real change, while they are all over the "slowdown in global climate change" as either an artifact of starting point or as statistically insignificant.

I think the truth on both is closer to "it's too early to tell."