Metric: Difference between revisions
Amwelladmin (talk | contribs) No edit summary Tags: Mobile edit Mobile web edit |
Amwelladmin (talk | contribs) No edit summary Tags: Mobile edit Mobile web edit |
||
| Line 12: | Line 12: | ||
An excellent page of resources on [[Goodhart’s law]] to be found [https://modelthinkers.com/mental-model/goodharts-law here]. | An excellent page of resources on [[Goodhart’s law]] to be found [https://modelthinkers.com/mental-model/goodharts-law here]. | ||
Regressive | *'''Regressive''': using a single metric as a proxy to measure a phenomenon that is actually [[multivariate]] — caused by several factors. Here [[Simpson ’s paradox]] is not your friend. | ||
*'''Extremal''': Where a the metric is a useful indicator under normal circumstances — [[Mediocristan]], but breaks down in extremes or unusual cases — [[Extremistan]]. Paging Messrs [[Black Scholes option pricing model|Black and Scholes]]. You know, using [[normal distribution]]s of independent events to model dependent events, like human behaviour of the market. | |||
*'''Causal''': the old [[correlation]] is not the same as [[causation]] chestnut. | |||
*'''Adversarial''': Substitute targeting the desired outcome — a senior tranche in a portfolio of mortgages that will not default in any circumstances — with one that is rated AAA. Hello, [[global financial crisis]]. | |||
{{Outsourcing}} | {{Outsourcing}} | ||
*[[Beware of shorthand]] | *[[Beware of shorthand]] | ||
*[[Multivariate]] factors | |||
Revision as of 09:16, 2 May 2023
|
The JC sounds off on Management
|
When everything about a people is for the time growing weak and ineffective, it begins to talk about efficiency. ... Vigorous organisms talk not about their processes, but their aims.
- — G. K. Chesterton, Heretics
“When a measure becomes a target, it ceases to be a good measure.”
The stock-in-trade of a middle manager and the management consultant she aspires to become.
Like a key performance indicator, a second-order derivative of actual performance calculated to allow non-experts to make cavalier management decisions, usually to reduce expenditure on — aka make redundant — the person performing that function.
To be contrasted with the ineffable, inarticulable skills that are provided by a subject matter expert.
Goodhart’s law
An excellent page of resources on Goodhart’s law to be found here.
- Regressive: using a single metric as a proxy to measure a phenomenon that is actually multivariate — caused by several factors. Here Simpson ’s paradox is not your friend.
- Extremal: Where a the metric is a useful indicator under normal circumstances — Mediocristan, but breaks down in extremes or unusual cases — Extremistan. Paging Messrs Black and Scholes. You know, using normal distributions of independent events to model dependent events, like human behaviour of the market.
- Causal: the old correlation is not the same as causation chestnut.
- Adversarial: Substitute targeting the desired outcome — a senior tranche in a portfolio of mortgages that will not default in any circumstances — with one that is rated AAA. Hello, global financial crisis.