pps43

9 ms·

61.
by pps43 6y ago
> the US had better invest heavily in education At this point, more money is not going to help. Education in the US is fundamentally broken. Combination of "no child is left behind" and standardized testing (because everything
62.
by pps43 6y ago
I would not call it a drawback. In regular regression you make an assumption that x is known exactly, and all errors are in y. In Deming regression you make another assumption, about the ratio of their variances (δ). If you change units, yo
63.
by pps43 6y ago
> that's really the most obvious cost function To me Deming regression is the most obvious. It actually took me a long time to realize that x~y and y~x are in most cases different lines.
64.
by pps43 6y ago
Applied statistics.
65.
by pps43 6y ago
Not really. It was tried and it turns out that the best strategy for the human is to just do what the computer suggests.
66.
by pps43 6y ago
Existing judicial system is calibrated for HUMINT. False accusations are costly, but rare. Using digital communications to discover criminals can accidentally sweep in many more innocents, who would then have to hire lawyers and carry all k
67.
by pps43 6y ago
Look at non-COVID-19 related deaths in a location with lockdown in effect, but with few COVID-19 deaths. Those should not be people who died from COVID-19 without a test, but if lockdown drives deaths up, there should be enough of them to b
68.
by pps43 6y ago
Twitter does not want to be a publisher, because then lots of people will sue it for libel and other nasty stuff in twits.
69.
by pps43 6y ago
CDC publishes weekly number of deaths from all causes: https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm
70.
by pps43 6y ago
You are extrapolating linear relationship between unemployment and deaths way outside the range of historical data on which it was estimated. Surely we would see some of the ~777k deaths already. The fact that we don't suggests that th
71.
by pps43 6y ago
To get oscillations, you need to shift the phase by 180 degrees. Capacitance alone (without some inductance) can only do 90.
72.
by pps43 6y ago
This question tests understanding of limitations of idealized circuits. Anyone with hands-on experience of shorting a capacitor knows about sparks and arcing that accompanies it.
73.
by pps43 6y ago
If inductance is zero, there would be no oscillations.
74.
by pps43 6y ago
A real-world analogy is a brick standing upright on the table. If you topple it and it falls, some of the potential energy is gone.
75.
by pps43 6y ago
That's one way to solve it. More generally, thinking of edge cases or simplifying the problem is a good problem-solving strategy.
76.
by pps43 6y ago
> Russians call everything a rocket anyway. RPG-7 is a rocket Bad example. РПГ is "ручной противотанковый гранатомёт", approximately "man-portable anti-tank grenade launcher". Its round, "ПГ-7" is "anti
77.
by pps43 6y ago
To produce a square wave, speaker cone should move with infinite speed.
78.
by pps43 6y ago
I'm not trying to convince you. I'm just saying that the data you quoted does not contradict the model I linked to. New cases in Iowa are essentially flat for a while now, which is consistent with Rt close to 1. https:/&#x2F
79.
by pps43 6y ago
Doubling of cumulative cases in 6-10 days, after adjusting for increased testing, is not that far off from Rt=0.94. Even Rt=1.1 still falls within the uncertainty interval.
80.
by pps43 6y ago
There are more confirmed cases because there's more testing.
81.
by pps43 6y ago
> the US epidemic is still large and still growing Rt is below 1 in all states except Minnesota and Nebraska [1]. So the US epidemic is still large, but already shrinking. With that said, when lockdowns are eased, Rt can increase again.
82.
by pps43 6y ago
Either that, or pool the data from all studies and analyze it together, almost certainly finding a statistically significant effect.
83.
by pps43 6y ago
What if there are 10 independent studies, all showing positive, but not statistically significant effect?
84.
by pps43 6y ago
Premature optimization. If you end up using a tool only once, buying the best quality tool for that is a waste of money. And you can rarely tell in advance which tool you'll be using often.
85.
by pps43 6y ago
Require masks in enclosed public places.
86.
by pps43 6y ago
PCR tests are asymmetric. Positive result almost certainly means that this person has the virus, but negative result might mean a number of things. Bad swab (less than 3000 virus copies), temporary remission (Korea has at least 160 such cas
87.
by pps43 6y ago
They don't have to always work well. If Alice's mask stops half the particles and Bob's mask stops half the particles, the chance of Alice infecting Bob is now 4x lower. 50% is not as good as 95%, but surely better than 0%. A
88.
by pps43 6y ago
> no one really knows Healthcare workers wear masks. I'm pretty sure they have a pretty good idea whether masks work.
89.
by pps43 6y ago
This paper? "I think the authors of the above-linked paper owe us all an apology. We wasted time and effort discussing this paper whose main selling point was some numbers that were essentially the product of a statistical error."
90.
by pps43 6y ago
Not rare. I shoot USPSA and Steel Challenge and have seen all kinds of revolver malfunctions in matches. Cylinder binding because bullets walked out under recoil, broken firing pin, cylinder falling out on the ground, timing issues (chamber

More ›