A recent study by Princeton economists Anne Case and Angus Deaton (a Nobel prizewinner) published in Proceedings of the National Academy of Sciences (http://ift.tt/1NmqVjk) found that 45-54 yo non-Hispanic males had experienced increasing rates of mortality during 1999-2013. As noted in the study, other broad mortality trends (other countries, other demographic groups, etc) show patterns of decreasing mortality.
An article in NY Times (http://ift.tt/1NAcVhL) captured some of the "holy cow!" reactions of other experts in the field. A theme is that somehow this hidden-in-the-data trend was missed by other experts.
I'm wondering what our actuarial group, particularly actuaries who study their clients' data for OASDI or large public plans, have observed. Had you already observed something similar? Are you now reviewing data and if possible slicing/dicing to see whether the Case/Deaton finding is replicated in client data?
I personally am willing to believe the published study, I guess, but I will remain a bit skeptical until I see these results replicated by someone else and/or confirmed in OASDI or public plan experience reports.
Who remembers the Reinhart Rogoff paper on the bad things that happen when a country's debt as a percent of GDP yada yada yada? Turned out that these famous people had spreadsheet errors that were undetected for years. The Case/Deaton study uses multiple data sets to tease out deeply hidden results. How consistently do these datasets define white or Hispanic or Asian? What happens when the person declines to make his own designation? How was in-migration handled? Out-migration?
Or maybe I just want to hear from actuaries regarding an important mortality study.
An article in NY Times (http://ift.tt/1NAcVhL) captured some of the "holy cow!" reactions of other experts in the field. A theme is that somehow this hidden-in-the-data trend was missed by other experts.
I'm wondering what our actuarial group, particularly actuaries who study their clients' data for OASDI or large public plans, have observed. Had you already observed something similar? Are you now reviewing data and if possible slicing/dicing to see whether the Case/Deaton finding is replicated in client data?
I personally am willing to believe the published study, I guess, but I will remain a bit skeptical until I see these results replicated by someone else and/or confirmed in OASDI or public plan experience reports.
Who remembers the Reinhart Rogoff paper on the bad things that happen when a country's debt as a percent of GDP yada yada yada? Turned out that these famous people had spreadsheet errors that were undetected for years. The Case/Deaton study uses multiple data sets to tease out deeply hidden results. How consistently do these datasets define white or Hispanic or Asian? What happens when the person declines to make his own designation? How was in-migration handled? Out-migration?
Or maybe I just want to hear from actuaries regarding an important mortality study.
Mortality of midlife white non-Hispanic Males