|Source: wm.edu - Larry Leemis|
|Source: www05.usu.edu - Juergen Symanzyk|
Actually, there is a big difference between an actuary and a statistician, but first some similarities:
-Both have professional societies that require adherence to a code of ethical conduct.
-Both are highly trained in mathematics and other technical analysis, but need to be effective in communications of technical concepts to lay persons.
-Both are professions that are the expert in a field of significant social consequence. For actuaries, we need only think of Social Security, Medicare, Medicaid and the Affordable Care Act, among many others. For statisticians, any field that requires making inferences from data analysis, such as whether a particular drug is safe and effective, whether eating jelly doughnuts causes heart disease, or what are the chances a particular individual or team will win a contest. With regard to climate change, scientists use the tools of statisticians to determine what may be happening and when, while actuaries use these results to develop insurance products and measure the estimated costs.
OK -The big difference is that actuaries must adhere to an extensive and detailed written code of professional standards (not just ethics) that govern methods and assumptions for specific calculations and communication of results. The Actuarial Standards of Practice now number 50 and apply to life, health, pension, and casualty areas of practice. In some cases, deviation from the standard is permitted, but only if communicated properly with the delivery of calculation results. Actuaries started up the interim Actuarial Standards Board in 1985 and have worked ever since to develop and refine their rules with new standards and amendments to existing standards as the world changes.
On the other hand, statisticians seem to be in a Wild West mode, only recently moving informally in the general direction of compiling their own standards.
Statisticians have their own unique challenge to the development of statistical standards. There seem to be two schools of thought that affect methodology in statistics - Bayesian and frequentist - that sometimes result in conflicting opinions among professionals. But that is hardly an excuse for the lack of comprehensive standards given that there is only one reality we are trying to model.
2016 is shaking out as the year of major developments for statistical guidance - yes, 30+ years after the actuaries:
-March 2016 ASA Releases Statement on Statistical Significance and P-Values. A brief readable summary of the next item.
-June 2016. The American Statistical Association journal The American Statistician publishes The ASA's Statement on P-Values: Context, Process, and Purpose by Wasserstein, Ronald L. and Nicole A. Lazar. Summary of ASA position recognizing a key problem in statistical methods as currently employed.
-June 2016 Ten Simple Rules for Effective Statistical Practice published in the open access section of PLOS Computational Biology. Informal high level guidance on best practice.
The ASA guidance to date counts as slight nudging of their profession compared with the comprehensiveness of the actuarial standards, which also come with an Actuarial Board of Counseling and Discipline for enforcement. Now, the statisticians and actuaries mostly work in differing environments with many actuaries constrained by the realities of the corporate world - their results affect decisions made by their insurance companies, their corporate clients, or government regulators. By way of contrast, so much of the work in statistics is research by those in academia, with pressure on finding meaningful results for publication, often influenced by external funding constraints.
One solution would be for the American Academy of Actuaries to cross professional lines and offer the American Statistical Association help to get them moving on their own comprehensive standards.
OK, back to that first question about the real difference between an actuary and a statistician, we should have tried this puzzler:
OK , so looks are deceiving, but you already knew that. So what. Well, unlike actuaries, statisticians are experiencing an existential crisis which calls into question the conclusions of many scientific studies of recent years, especially in, but not limited to psychology. The "p-hacking" or "replication" crisis in statistics bears similarities to the current journalistic crisis which finally calls into question the traditional model of impartial journalism that gives equal weight to opposing sides and defines objectivity as the mid-point between two sides arguing.