nickforr

I guess traditionally ‘middle class’ referred to the type of occupation (e.g. non-manual) which was typically associated with being better off.

But it’s not obvious that the standard of living associated with being better off would ever have been near the 90th, let alone 90th percentile of salaries?

Not convinced that sending children to private school would ever have been seen as a ‘lower middle class’ expectation.

But I’m also not convinced that the markers you describe are not available to someone at the 75th percentile of income, say, let alone to people at the 95th percentile. Now the luxuriousness of those markers may not be at the level marketed in glossy brochures etc but isn’t that an issue with unrealistic expectations?

I think some perspective is missing if you’re describing this as “pushing anyone down that attempts to break into the middle class” when the cutoff points (when the step changes occur, as you’re correct to note) come into effect at around something like the 95th percentile of UK salaries.

https://thesalarysphere.com/blog/average-salary-uk/

If you write the R code using vectorised operations it’s significant orders of magnitude faster (148 seconds to 1.5 milliseconds on posit.cloud).

The nested loop becomes: j <- sum((seq_len(100000) - 1) %% u) a <- a + j + r

A large number of statistical packages in R make use of C++ for their core algorithms. See here, for example: https://cran.r-project.org/web/packages/Rcpp/index.html

With apologies if this breaks guidelines: https://hymans.current-vacancies.com/Jobs/Advert/3525353?cid...

Just to pick one example, do you see something like state-imposed food standards as a 'hurdle' added by bureaucrats? What about housing construction standards? One person's hurdle can be another person's safety net?

If you want to sell in two different countries don’t you have to comply with two different legal codes and support both of them?

How does the PM know better than devs about whether or not they’re bringing the right information and not inadvertently summarising something incorrectly that’s technically important? Gatekeepers are generally inefficient longer-term.

Surely the context for the time series matters? Not sure good outcomes will follow from people blindly applying black boxes.

And if the revenue tanked would you accept half your salary, say, or would you expect a floor to be applied?

But without some statistical knowledge, isn’t there a risk of a lack of understanding about the robustness of “what works”?

Why would you not want business people to understand the technical constraints? Maybe easier in the short term to make the decision for them but longer term that just builds up barriers and an “us” and “them” culture, which becomes very unhealthy.

Worth looking through here and related links: https://statmodeling.stat.columbia.edu/2020/04/19/fatal-flaw...