Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

That's too noisy given inflation is 6.6%.


Real GDP

We've only recently seen nominal GDP declines in 2008 and 2020.


I know it's real GDP. I'm saying this number is noisier when inflation is high because you only ever measure nominal GDP and then deflate.


Well what don't you believe because you're using 6.6% inflation as your reason for skepticism, but that's what GDP is deflated by?


They believe the error bar on that inflation number must be pretty wide, presumably.


But by that logic, error bars increased because the number went up? Idk if that’s how statistics works.


Errors in counting generally go as square root of the count. So yes, as number goes up so does the deviation.


Inflation readings being higher have nothing to do with a higher count.

Exact same amount of data points for inputs as 2% inflation.


It’s not about the amount of measurement points, uncertainties in dollar amounts tend to scale as sqrt. When you error propagate that you see that inflation uncertainty scales with inflation.

Very simple model

Inflation=v1/v2

D(inflation)=sqrt((D(v1)/v2)^2+(v1/v2^2 D(v2))^2)

Suppose

D(v1)=c sqrt(v1)

D(v2)=c sqrt(v2)

D(inflation)=c sqrt(v1/v2^2+v1^2/v2^3)= c sqrt(inflation+inflation^2)/sqrt(v2)

So can be seen that the leading term for uncertainty in inflation is linear with inflation.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: