Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Forecasts need to have error bars

andrewpwheeler.com

Post reply on HN
Print view View original

141–150 of 165 posts

Re: Forecasts need to have error bars

#141

Post by gh02t »

marcyb5st wrote: post #137
gh02t wrote: post #72

Earlier quoted context omitted.

You can demand error bars but they aren't always possible or meaningful. You can more or less "fudge" some sort of normally distributed IID error estimate onto any method, but that doesn't necessarily mean anything. Generating error bars (or generally error distributions) that actually describe the common sense idea of uncertainty can be quite theoretically and computationally demanding for a general nonlinear model…

Well, in reality tools like Tensorflow probability can help you model both aleatoric and epistemic uncertainty with probabilistic layers that have learnable priors and posteriors. The issue there is that for the average ML person might not have the required math skills to model the problem in these terms. For instance, if you look at https://blog.tensorflow.org/2019/03/regression-with-probabil... until the case 4 it'…

Yeah I've used those methods and am a fan, though they are far from perfect. For one thing they're somewhat invasive methods to implement and they still require you to formulate a likelihood function to varying degrees; a task which is not always possible up front. I've also had issues with getting them to converge during training when using them. They also sometimes don't estimate uncertainty correctly, particularly if you make a mistake modeling the likelihood.

I guess my point is, there is no silver bullet. Adding defensible uncertainty is complicated and problem specific, and comes with downsides (often steep).

Re: Forecasts need to have error bars

#142

Post by gh02t »

kqr wrote: post #74
gh02t wrote: post #72

Earlier quoted context omitted.

You can demand error bars but they aren't always possible or meaningful. You can more or less "fudge" some sort of normally distributed IID error estimate onto any method, but that doesn't necessarily mean anything. Generating error bars (or generally error distributions) that actually describe the common sense idea of uncertainty can be quite theoretically and computationally demanding for a general nonlinear model…

I'm going to sound incredibly subjectivist now, but... the human running the model can just add error bars manually. They will probably be wide, but that's better than none at all. Sure, you'll ideally want a calibrated estimator/superforecaster to do it, but they exist and they aren't that rare. Any decently sized organisation is bound to have at least one. They just need to care about finding them.

Even subjectively, on what basis would they generate uncertainties that at least keeps some grounding in reality? Any human generation would be ad hoc and likely very wrong, humans are notoriously awful at estimating risk and I'd argue by extension uncertainties with any consistency. And that's not even considering how one would assign an uncertainty to some huge model with 350 wacko features trained on 40 million examples. Lastly, models don't necessarily attend to the same details a human does so even if a human is able to slap an uncertainty on a prediction based on their own analysis that doesn't mean it's representing the uncertainty of what the model based its decision on.

I do think having people in the loop is a very important aspect, however, and can provide an important subjective complement to the more mathematically formulated idea of uncertainty. I don't care if the model I'm using provides the most iron clad and rigorous uncertainties ever, I'm still going to spot check it and play with it before I consider it reliable.

Re: Forecasts need to have error bars

#143

Post by ISL »

bluecheese452 wrote: post #138
ISL wrote: post #99

Earlier quoted context omitted.

A position espoused by Bill Phillips [1], and to which I now adhere: "You should be willing to take either side of the bet that confidence interval implies." (paraphrasing; he says it better). For a concrete example, with a 95% confidence interval, you should be as willing to accept the 19:1 odds that the true value is outside the interval as you are the 1:19 odds that the true value is inside the interval. Aside fro…

Expected value does not equal utility. I am not willing to mortgage my 1 million dollar house for a 1 in a 1000 shot at a billion.

Then one should be very careful when assigning 99.9% confidence intervals.

Re: Forecasts need to have error bars

#144

Post by bluecheese452 »

ISL wrote: post #143
bluecheese452 wrote: post #138

Earlier quoted context omitted.

Expected value does not equal utility. I am not willing to mortgage my 1 million dollar house for a 1 in a 1000 shot at a billion.

Then one should be very careful when assigning 99.9% confidence intervals.

The two are unrelated.

Re: Forecasts need to have error bars

#145

Post by cubefox »

bo1024 wrote: post #80
cubefox wrote: post #68

Earlier quoted context omitted.

"Credible interval": https://en.wikipedia.org/wiki/Credible_interval

No, predictive interval is more precise, since we are dealing with predicting an observation rather than forming a belief about a parameter.

What's a predictive interval?

Re: Forecasts need to have error bars

#146

Post by borg16 »

Reminds me of this paper[1]

> An illusion of predictability in scientific results: Even experts confuse inferential uncertainty and outcome variability

> Traditionally, scientists have placed more emphasis on communicating inferential uncertainty (i.e., the precision of statistical estimates) compared to outcome variability (i.e., the predictability of individual outcomes). Here, we show that this can lead to sizable misperceptions about the implications of scientific results. Specifically, we present three preregistered, randomized experiments where participants saw the same scientific findings visualized as showing only inferential uncertainty, only outcome variability, or both and answered questions about the size and importance of findings they were shown. Our results, composed of responses from medical professionals, professional data scientists, and tenure-track faculty, show that the prevalent form of visualizing only inferential uncertainty can lead to significant overestimates of treatment effects, even among highly trained experts. In contrast, we find that depicting both inferential uncertainty and outcome variability leads to more accurate perceptions of results while appearing to leave other subjective impressions of the results unchanged, on average.

[1] https://www.microsoft.com/en-us/research/publication/an-illu...

Re: Forecasts need to have error bars

#147

Post by kqr »

gh02t wrote: post #142
kqr wrote: post #74

Earlier quoted context omitted.

I'm going to sound incredibly subjectivist now, but... the human running the model can just add error bars manually. They will probably be wide, but that's better than none at all. Sure, you'll ideally want a calibrated estimator/superforecaster to do it, but they exist and they aren't that rare. Any decently sized organisation is bound to have at least one. They just need to care about finding them.

Even subjectively, on what basis would they generate uncertainties that at least keeps some grounding in reality? Any human generation would be ad hoc and likely very wrong, humans are notoriously awful at estimating risk and I'd argue by extension uncertainties with any consistency. And that's not even considering how one would assign an uncertainty to some huge model with 350 wacko features trained on 40 million ex…

> on what basis would they generate uncertainties that at least keeps some grounding in reality?

By having their forecasts continuously evaluated against outcomes. If someone can show me they have a track record of producing calibrated error bars on a wide variety of forecasts, I trust them to slap error bars on anything.

> even if a human is able to slap an uncertainty on a prediction [...] that doesn't mean it's representing the uncertainty of what the model based its decision on.

This sounds like it's approaching some sort of model mysticism. Models don't make forecasts, humans do. Humans can use models to inform their opinion, but in the end, the forecast is made by a human. The human only needs to put error bars on their own forecast, not on the internal workings of the model.

Re: Forecasts need to have error bars

#148

Post by jrumbut »

eru wrote: post #114
jrumbut wrote: post #106

Earlier quoted context omitted.

> Also assuming normal distribution, I think? 95% is 95% regardless of the distribution. > I would like to build some edge into my bets. If a reader takes both sides of your example, they would be come out exactly even. You can imagine yourself being equally unhappy to take either side of the bet, if that's easier than imagining yourself being happy to take either side. It is for me, which is probably something to br…

> 95% is 95% regardless of the distribution. Standard deviation away from the mean don't correspond to the same percentiles for all distributions, or do they? If you want to be (almost) independent of distribution, you need Chebyshev's inequality. But that one is far weaker. > Its practical usage is similar to the 68–95–99.7 rule, which applies only to normal distributions. Chebyshev's inequality is more general, sta…

> Standard deviation away from the mean don't correspond to the same percentiles for all distributions, or do they?

If they had said a two standard deviation interval then you would have needed to know the distribution, but they said 95% which gives you all the information you need to make the bet.

Re: Forecasts need to have error bars

#149

Post by jmalicki »

nradov wrote: post #31
figassis wrote: post #23

Earlier quoted context omitted.

So is it really science? These are concepts from stats 101. And the reasons and need, and the risks of not having them are very clear. But you have millions being put into models without these pre-requisites, and being sold to people as solutions, and waved away as "if people buy is it's bc it has value". People also pay fraudsters.

Mostly not. Very few data "scientists" working in industry actually follow the scientific method. Instead they just mess around with various statistical techniques (including AI/ML) until they get a result that management likes.

Science is just fooling around with data until you get a result a journal reviewer likes.

Re: Forecasts need to have error bars

#150

Post by hgomersall »

8organicbits wrote: post #134
hgomersall wrote: post #118

Earlier quoted context omitted.

I'm not sure I see the distinction. Would you mind clarifying?

Model: water freezes below 0° C. Input: temperature is measured at -1° C. Prediction: water will freeze. Actual: water didn't freeze. Actual temperature: 2° C. The model isn't broken, it gives an incorrect result because of input error.

Well I'd say the model is broken because it didn't capture the uncertainty in the measurements.
Post reply on HN