It actually does have practical applications for me, because it will be part of my calculations. I don’t know whether I should have any preference for the distribution of utility over my lifetime at all, before I consider things like uncertainty and opportunity cost. Does this mean you would say the answer is no?
AstraSequi
Karma: 90
I can think of example where I behaved both ways, but I haven’t recorded the frequencies. In practice, I don’t feel any emotional difference. If I have a chocolate bar, I don’t feel any more motivated to eat it now than to eat it next week, and the anticipation from waiting might actually lead to a net increase in my utility. One of the things I’m interested in was whether there’s anyone else who feels this way, because it seems to contradict my understanding of discounting.
That assumption is to make time the only difference between the situations, because the point is that the total amount of utility over my life stays constant. If I lose utility during the time of the agreement, then I would accept a rate that earns me back an amount equal to the value I lost. But if I only “want” to use it today and I could use it to get an equal amount of utility in 3 months, then I don’t have a preference.
Thanks for that – the point that I’m separating out uncertainty helped clarify some things about how I’m thinking of this.
So is time inconsistency the only way that a discount function can be self-inconsistent? Is there any reason other than self-inconsistency that we could call a discount function irrational?
Second, with respect to “my intuition is not to discount at all”, let’s try this. I assume you have some income that you live on. How much money would you take at the end of three months to not receive any income at all for those three months? Adjust the time scale if you wish.
If I received an amount equal to the income I would have gotten normally, then I have no preference over which option occurs. This still assumes that I have enough savings to live from, the offer is credible, there are no opportunity costs I’m losing, no effort is required on my part, etc.
In general, you can think of discounting in terms of loans. Assuming no risk of default, what is the interest rate you would require to lend money to someone for a particular term?
This is the same question, unless I misunderstood. I do have a motivation to earn money, so practically I might want to increase the rate, but I have no preference between not loaning and a rate that will put me in the same place after repayment. With my assumptions, the rate would be zero, but it could increase to compensate—if there’s an opportunity cost of X, I’d want to get X more on repayment, etc.
I have some questions on discounting. There are a lot, so I’m fine with comments that don’t answer everything (although I’d appreciate it if they do!) I’m also interested in recommendations for a detailed intuitive discussion on discounting, ala EY on Bayes’ Theorem.
Why do people focus on hyperbolic and exponential? Aren’t there other options?
Is the primary difference between them the time consistency?
Are any types of non-exponential discounting time-consistent?
What would it mean to be an exponential discounter? Is it achievable, and if so how?
What about different values for the exponent? Is there any way to distinguish between them? What would affect the choice?
Does it make sense to have different discounting functions in different circumstances?
Why should we discount in the first place?
On a personal level, my intuition is not to discount at all, i.e. my happiness in 50 years is worth exactly the same as my happiness in the present. I’ll take $50 right now over $60 next year because I’m accounting for the possibility that I won’t receive it, and because I won’t have to plan for receiving it either. But if the choice is between receiving it in the mail tomorrow or in 50 years (assuming it’s adjusted for inflation, I believe I’m equally likely to receive it in both cases, I don’t need the money to survive, there are no opportunity costs, etc), then I don’t see much of a difference.
Is this irrational?
Or is the purpose of discounting to reflect the fact that those assumptions I made won’t generally hold?
The strongest counterargument I can think of is that I might die and not be able to receive the benefits. My response is that if I die I won’t be around to care (anthropic principle). Does that make sense? (The discussions I’ve seen seem to assume that the person will be alive at both timepoints in any case, so it’s also possible this should just be put with the other assumptions.)
If given the choice between something bad happening now and in 10 years, I’d rather go through it now (assume there are no permanent effects, I’ll be equally prepared, I’ll forget about the choice so anticipation doesn’t play a role, etc). Does that mean I’m “negative discounting”? Is that irrational?
I find that increasing the length of time I anticipate something (like buying a book I really want, and then deliberately not reading it for a year) usually increases the amount of happiness I can get from it. Is that a common experience? Could that explain any of my preferences?
I think the value of a Wikipedia pageview may not be fully captured by data like this on its own, because it’s possible that the majority of the benefit comes from a small number of influential individuals, like journalists and policy-makers (or students who will be in those groups in the future). A senator’s aide who learns something new in a few years’ time might have an impact on many more people than the number who read the article. I’d actually assign most of my probability to this hypothesis, because that’s the distribution of influence in the world population.
ETA: the effects will also depend on the type of edits someone makes. Some topics will have more leverage than others, adding information from a textbook is more valuable than adding from a publicly available source, and so on.
This can be illustrated by the example of evolution I mentioned: An evolutionary explanation is actually anti-reductionist; it explains the placement of nucleotides in terms of mathematics like inclusive genetic fitness and complexities like population ecology.
This doesn’t acknowledge the other things explained on the same grounds. It’s a good argument if the principles were invented for the single case you’re explaining, but here they’re universal. If you want to include inclusive genetic fitness in the complexity of the explanation, I think you need to include everything it’s used for in the complexity of what’s being explained.
Sure, this experiment is evidence against ‘all fat, tired people with dry hair get better with thryoxine’. No problem there.
Okay, but you said it was evidence in favor of your own hypothesis. That’s what my question was about.
Yes, it is kind of odd isn’t it? One of the pills apparently made them a bit unwell, and yet they couldn’t tell which one. I notice that I am confused.
Suppose they’re measuring on a 10-point scale, and we get ordered pairs of scores for time A and time B. One person might have 7 and 6, another has (4,3), another has (5,6), then (9,7), (7,7), (4,5), (3,2)...Even if they’re aware of their measurements (which they might not be), all sorts of things affect their scores and it’s unlikely that any one person would be able to make a conclusion. You’re basically asking an untrained patient to draw a conclusion from an n of 1.
But that’s awful! Once, there was a diagnostic method, and a treatment that worked fine, that everyone thought was brilliant. Then they invented a test, which is very clever, and a good test for what it tests, and the result of that is that lots of people are ill and don’t get the treatment any more and have to suffer horribly and die early.
There are several assumptions here that I think are probably incorrect, the biggest being the causal link between introducing the test and people suffering. But what I described before is just the application of reductionism to better distinguish between disease states based on their causal mechanism.
If that’s normal then there’s something badly wrong with normal. A new way of measuring things should help!
Sometimes, but replacing an objective measurement with a subjective one isn’t usually a step forward.
Seriously, if ‘start off with low doses and keep raising the dose until you get a response’ is inaccessible to testing, then something is broken.
Problems with this include: you can’t justify the parameters of the dose increase, you still have to agree on how to measure the response, and you also have a multiple testing issue. It isn’t inaccessible, but it’s a complication (potentially a major one), and that’s just in the abstract. Practically, in any one situation there might be another half dozen issues that wouldn’t be apparent to anyone who isn’t an expert.
But in fact, just ‘low basal metabolic rate in CFS’ would be good evidence in favour, I think. We can work out optimal treatments later.
Not knowing anything about the subject, I would expect to observe a low basal metabolic rate in CFS regardless of its ultimate cause or causes.
At that point, we’re all post-modernists aren’t we? The truth is socially determined.
No, it just means we put very little weight on individual studies. We don’t pay much attention to results that haven’t been replicated a few times, and rely heavily on summaries like meta-analyses.
Science is not unreliable...
You’re talking about the overall process and how science moves in the direction of truth, which I agree with. I’m talking on the level of individual papers and how our current best knowledge may still be overturned in the future. But you can leave out “just like..wisdom” from the paragraph without losing the main points.
There’s at least a possibility here that medical science is getting beaten hollow by chiropractors and quack doctors and internet loonies, none of whom have any resources or funding at all.
The alt med people have a lot of funding. It’s a multi-billion-dollar industry.
Even the possibility is enough to make me think that there’s something appallingly badly wrong with the methods and structure of medical science.
A few things, not just one, but it’s the best we have at the moment.
This open-access article discusses some of the issues in cancer research.
In most ways biology is intermediate between the hard and soft sciences, with all that implies. It’s usually impossible to identify all the confounders, most biologists are not trained in statistics, experiments are complex and you can get different results from slight variations in protocol, we’re trying to generalize from imperfect models, many high-profile results don’t get tested by other labs, … all these factors come together and we get something that people call a “replication crisis.”
If none of the patients had had any sort of thyroid problem, I’d have expected it to be equally bad for everyone.
I’m talking about conservation of expected evidence. If X is positive evidence, then ~X is negative evidence. An experiment only supports a hypothesis if it was possible for it to come out another way that refutes it. And if an experiment that could have supported the hypothesis actually didn’t, then it’s evidence against.
What makes me think that they felt bad on thyroxine is table 2, where all the ‘self-reported’ psychological scores have got worse from thyroxine. In particular p=0.007 for the decline in Vitality. Since, as you point out, they really didn’t know which was which, it’s hard to see how they could have faked that.
Terminology then. When you said “Thyroxine is very strongly disliked by the healthy controls (they could tell it from placebo and hated it),” it suggests they could identify the active treatment.
Absolutely this treatment is harmful to healthy people.
The people in the study had symptoms. Even if you think their symptoms were mild or unrepresentative, you shouldn’t call them healthy. It’s fair to extend the conclusion to cover people without those symptoms, but I think that’s an important difference.
Yes, but that does mean that anything that needs careful dose control will get rejected.
It’s more that you need an easily followed protocol. Anything else, especially anything subjective, is unlikely to be practically feasible, and will probably not be reproducible.
The TSH test replaced that around 1970. But they never seem to have checked that clinical and biochemical diagnoses detected the same things, and after that there was the slow emergence of all sorts of nasty diseases that look very like hypothyroidism in the clinical sense but have normal TSH.
This is normal. Clinical presentations often have many causes, which makes it almost impossible to progress. Eventually we break them down based on their causal mechanisms so we can treat them individually. Each time we find a new cause, some of the cases will be left unexplained.
These are the only ones I can find through google scholar / pubmed. That in itself is really surprising and one of the things I can’t explain! Why has such an obvious thing not been ruled out?
There are a lot of interesting hypotheses competing for resources, and we have to decide which ones are worth considering. I can’t say what the reason might be here, but there are a lot of possibilities. For example, it might not be possible to design a study like the one you want that could effectively answer the question.
Really? Forty years of experience in treating patients is less valuable than a single anecdote published in a journal? Really?
Yes. Expert opinion (i.e., the opinion of individual experts, not expert consensus) is the lowest level because you can find an expert to support pretty much any proposition that isn’t obviously ridiculous, and sometimes even if it is. In fact, this is true higher in the hierarchy as well, which is why we use syntheses of evidence so much. I can’t stress this enough: in biology, you can use peer-reviewed evidence to make plausible arguments for arbitrary hypotheses.
All the rest of it is anecdotal, from alternative sources, but there’s a mountain of it.
The point of evidence-based medicine is that perceptions are unreliable. That includes the perceptions we call clinical experience (which once said that bloodletting was an important medical treatment). Keep in mind that doctors aren’t scientists and usually don’t even qualify as experts. EBM is unreliable too, but less so, just like science is unreliable but is still better than ancestral wisdom.
The TSH test ruling out hypothyroidism is expert opinion. Its reliability is unfounded dogma.
This sounds like you’re saying the TSH test doesn’t actually measure TSH, but I think you mean to say you disagree with the conclusions that it’s used for. But since hypothyroidism is defined as low thyroid hormone levels, some of this will be a dispute over definitions.
I can’t find any evidence for it as the sole measure of thyroid system function at all.
I don’t think anyone who understands it would say it is. It measures TSH levels, and the question is what we do with that measurement. But we’re often limited by what we’re able to (easily) measure, and it might even be the only objective measurement we have.
Why is the Pollock trial evidence supporting your hypothesis? What outcome from the trial would you have considered to be evidence against it?
Also, what part suggests that the healthy controls could distinguish the treatment from placebo? From Table 4, it seems that the reverse is true.
At first glance, the results from that study look like straightforward evidence that this treatment is actively harmful. I’d also point out that RCTs need to be standardized across patients. I can’t say whether the inclusion criteria should have been different, but choosing a single dose is normal procedure. There are always better options, but it’s a weak argument on its own, in part because it can be applied in almost any circumstances.
Everyone who’s ever tried fixing the clinical diagnosis of hypothyroidism with any kind of thyroid therapy either seems to think it works, or hasn’t written about it on the internet or in the medical literature.
I admit I’m not an endocrinologist, but from what I’m reading I don’t think there is any recognized clinical diagnosis of hypothyroidism. The TSH test is the gold standard. That would suggest those who talk about it are primarily cranks and such.
That’s a big claim. I’m making it in bold on Less Wrong. I expect someone to turn up some evidence against it. I would love to see that evidence.
Less Wrong might not be the best place for this, since there aren’t many biologists here. You have the burden of proof (i.e., the prior for arbitrary hypotheses is very low), so you shouldn’t be asking other people to disprove it. Could you summarize your support for this claim? Are these the only two peer-reviewed articles?
Assuming he’s not just making up his data it’s hard to explain his results.
There are lots of ways that data can be wrong without being made up. 90% of medical research findings are false, etc.
This depends on what kind of unfalsifiability you want. There are at least four kinds.
unfalsifiable with current resources (Russell’s teapot)
unfalsifiable because of moving goalposts
unfalsifiable because the terms are incoherent or undefined (“not even wrong”)
unfalsifiable in principle
No empirical claim is unfalsifiable in principle (i.e. without resource limitations, moving goalposts, or logical incoherency). Claims that involve violations of physical law come the closest, but require us to assume 100% confidence in the law itself. For a non-empirical claim to be unfalsifiable, empirical consequences of the claim have to be impossible, which ultimately requires you to eliminate them by definition. I think you’re trying to find an example of the fourth meaning when most people who talk about unfalsifiability are thinking about one of the others.
Maybe “Is accurate enough that it doesn’t change our answer by an unacceptable amount”? The level of accuracy we want depends on context.
How would you measure the accuracy of a model, other than by its probability of giving accurate answers? “Accurate” depends on what margin of error you accept, or you can define it with increasing penalties for increased divergence from reality.
I don’t think I need too much data to assign broadly negative values to lives that are unusually brutish, nasty and short compared to either non-existence or a hypothetical natural existence.
I don’t think you can make that decision so easily. They’re protected from predators, well-fed, and probably healthier than they would be in the wild. (About health, the main point against is that diseases spread more rapidly. But farmers have an incentive to prevent that, and they have antibiotics and access to minimal veterinary treatment.)
‘no pig’ > ‘happy pig + surprise axe’
This leads me to conclusions I disagree with—like if a person is murdered, then their life had negative value.
Another way to generalize 4 is
Always correct your probability estimates for the possibility that you’ve made an incorrect assumption.
I don’t think “changes the issue” is the best way to say this, because there is always a probability that your model won’t work even if it doesn’t say something is impossible.
I don’t know about this being a category error though. I think “map 1 is accurate with respect to X” is a valid proposition.
I would add the reverse of #3: “There is evidence for it” doesn’t mean much on its own either, for the same reasons.
My sympathies for your loss.
In the tradition of “making up numbers and doing Fermi estimation is better than making up answers,” I would focus on the history. The frequency of past outcomes is always a good place to start (I think that’s in the Sequences somewhere) since there’s no need to consider causality, only frequency and genetic distance. An example:
Simplify and assume the cause is genetic (which will overestimate the probability; environmental or shared genetic-environmental has more randomness and will have occurrence closer to the population average). What is the total number of siblings for yourself and your spouse, including both of you, and how many stillbirths were there? Add your children to the number, including the one stillbirth, and weight those double because they’re the generation you want to know about. Calculate the percentage, then increase it by 5-10% as a crude correction for the assumption of a genetic cause. This is my estimate before you start thinking about causality.
Other things: If V is your son from a different relationship, his genetic distance is further so I would give him normal weight instead of double, but if L has other children I would still double them since the mother’s genetics are probably more important. Optionally add any of your siblings’ children, but weight them by half due to greater genetic distance. Check what percentage of stillbirths are genetic vs environmental, which could be used that to make a better correction than 5-10%. To avoid the multiple comparisons problem, make these choices before doing the analysis and commit not to change them.
Disclaimers: I am not a doctor or genetic counselor, and this is not medical advice. This is a superficial analysis written at 5am with the first few ideas I thought of, based on my unreliable intuitions about what sort of estimates might work. This sort of estimate is a lot weaker than direct evidence like the BMJ meta-analysis. I take no responsibility for any decisions that anyone makes...etc.
PS: you should probably assume the disclaimers apply to anything you read here. Also, I think another reason doctors avoid giving probabilities is that there can be legal consequences, especially if they’re misinterpreted.
To me, that’s sort of like saying “don’t worry, when I said 2+2=5, I was being informal.”
Very true. This is something I’ll try to change.
I think this is a special case of the problem that it’s usually easier for an AI to change itself (values, goals, definitions) than for it to change the external world to match a desired outcome. There’s an incentive to develop algorithms that edit the utility function (or variables storing the results of previous calculations, etc) to redefine or replace tasks in a way that makes them easier or unnecessary. This kind of ability is necessary, but in the extreme the AI will stop responding to instructions entirely because the goal of minimizing resource usage led it to develop the equivalent of an “ignore those instructions” function.