Talk:Risk/Archives/2017

Definition of risk

The definition of risk given here (potential of gaining or losing something of value) is idiosyncratic. It has emerged only in the last 25 years through the specific discipline of enterprise risk management (ERM) and is contested within that group. E.g., see current debates in the ISO 31000 community.

Historically, and in science, engineering, insurance and medicine, risk is a combination of the likelihood and severity of an unwanted event or outcome. Note that "combination" does not necessarily imply an arithmetic product. I.e., risk can be left as a 2-dimensional vector (as it is in aviation) rather than being expressed as a scalar value.

3a : "the chance of loss or the perils to the subject matter of an insurance contract; also : the degree of probability of such loss"

Risk‘s Latin root means danger. The 1828 American Dictionary of the English Language says risk signifies a degree of hazard or danger. In Google’s data, the top 25 two-word collocations starting with “risk” all associate risk with cost or loss, e.g., aversion, alerts, assessment, mitigation.

The notion of "positive risk" confuses risk-reward calculus with the concept of risk alone.

[I am unfamiliar with this process and am attempting to comply with guidelines.]

Bstorage (talk) 07:12, 10 December 2016 (UTC) William Storage

If by “idiosyncratic”, you mean that inclusion of “gaining of” makes the definition of risk pecular to a narrow or new usage, I agree. I am familiar with the concept that risk may be defined to include gain or benefits (e.g., positive risk), but I concede that it is a minority concept.

As for complying with guidelines, I would note that I don’t see that the [http://iosrjournals.org/iosr-jbm/papers/Vol16-issue3/Version-3/K016338389.pdf present citation] really supports the present definition “Risk is the potential of gaining or losing something of value.” The cited article really only concerns loss. The simple fix is to just delete “gaining or”. Alternatively, one should find a citation that actually uses that definition. However, as demonstrated by the real problem with the article, there is really very wide usage of the term, and so, the real challenge in editing the introductory paragraph is to accommodate the range of usage, and included in that introduction could be mention (with citation) of “positive risk”.

Presently, I think the article covers far too much ground, IMO a result of people from several disparate professions contributing piecemeal. If I could “throw pixie dust in the air”, the article would be majorly revised. Structurally, it is flawed, content doesn’t necessarily follow headings. Under the present title “Risk”, this should really be more of an uber-disambiguation article (OK, parts of it are), treating the various applications of risk in generalities only, and linking to several “Main Articles”.

IveGoneAway (talk) 15:09, 12 December 2016 (UTC) 16:53, 12 December 2016 (UTC)

“Note that "combination" does not necessarily imply an arithmetic product. I.e., risk can be left as a 2-dimensional vector (as it is in aviation) rather than being expressed as a scalar value.”

Actually, in many areas I have studied, risk is an arithmetic product, roughly. Observe that risk is generally quantized in units of value per time. In Aviation, the fatality impact of Level B vs. Level A effects is roughly a factor of 100 (1 or few deaths vs. hundreds of deaths), and so, the relative acceptable probabilities of the effects also show a factor of 100 (10^-7 vs. 10^-9) (see ARP4761) concluding that, roughly, one fatality per 10 million flight hours is acceptable risk. While AS9100 is also an aviation risk standard, the focus has been business risk, not safety (The 2016 revision introduced a Product Safety clause, effectively a brief functional safety addendum). Safety is considered, but in terms of business impact of accidents. It is practical, if roughly, to assess specific risks in terms of money loss per year; and from this assessment prioritize risk actions; e.g., recognized $1000 losses that happen 100 times per year should have the same priority as $10,000,000 losses that happen once in a hundred years (both are $100,000/year risks.) Similarly, it is not effective to implement a $10,000/year process to control a $100/year risk. Other disciplines make similar comparisons in their own rough units of risk; there is risk of physical injury related to staying physically fit, but that is compared to certain physical injury resulting from being sedentary. IveGoneAway (talk) 16:33, 12 December 2016 (UTC)