I used to do some data mining research many years ago, and has been wondering if the value premium is susceptible to overfitting. The research in this episode took a nice approach and demonstrated it. Now I wonder if the same approach can be applied to other empirical work to re-examine data overfitting impact. Great work and great episode!
Hasler doesn't show the value factor doesn't exist; rather it might be smaller than originally thought. It still appears statistically significant though.
@@alankoslowski9473 yeah that’s what I thought overfitting means - producing a model maximizing performance on existing observations but not general enough for unseen data. A more general model would have shown a smaller value premium.
@@michaelspiano2022 Researchers have to make assumptions when examining historical data. I don't think Hasler is arguing his assumptions are necessarily better or more reasonable than FF, but rather they're just as reasonable and often it's uncertain what the most reasonable assumptions are.
Great content as always. Your content continues to empower so many people. Thanks to both of you and the whole team that helps get these episodes published :)
1. Hedging against inflation actually equals timing the market and is never a good idea. Just keep to your strategy. 2. Climate should not be part of your investment strategies. Just diversify and keep costs low. Live green, invest smart. 3. Value investing might give more returns over really long terms, but to me it seems not worth the risk. Sometimes it takes a decade or longer for the value factor to show up. What if my horizon of 25 years is too short for the value factor to emerge when tilting to value? Second, there is no good ETF available for me to even try this. I stick to a globally diversified portfolio and with that almost guarantee great results. Edit: Great video and discussion and congrats on your success on the website and in general. You tought me and some friends a lot of how to invest smart!
Does this change your view on value being a robust factor? And that a small-cap value tilt is still optimal for those willing to take on more complexity and risk?
Fair question, though he doesn't argue the value factor doesn't exist. He doesn't argue his assumptions are necessarily better, rather there are any number of reasonable, but arbitrary assumptions made in financial research. Even if it is smaller, the value premium still seems statistically significant.
I subscribe to Professor Andrew Lo's adaptive views of the markets so though the value factor may have taken a hit for decades, capitalism's mean reverting characteristics should suggest a resurgence of value but irritatingly, the question is when. With that being said, instead of trying to time and ride the possible wave of value resurgence, I'll just stick to my market cap weighted index funds.
I tried to create an account in the Rational Reminder Community and was rejected because of "Suspicious IP". I live in Brazil, so that's probably why, but I am a legit fan, I have messages in Ben's videos from years ago
![I.N "HALLUCINATION" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/n5B5q1Hwt_U/mqdefault.jpg)
![BLACK BAG - Official Trailer [HD] - Only in Theaters March 14](http://i.ytimg.com/vi/Du0Xp8WX_7I/mqdefault.jpg)
I used to do some data mining research many years ago, and has been wondering if the value premium is susceptible to overfitting. The research in this episode took a nice approach and demonstrated it. Now I wonder if the same approach can be applied to other empirical work to re-examine data overfitting impact. Great work and great episode!
Now that we discovered the “chance”, we must re-examine all major papers, especially positively skewed ones.
This episode blew my mind
Hasler doesn't show the value factor doesn't exist; rather it might be smaller than originally thought. It still appears statistically significant though.
@@alankoslowski9473 yeah that’s what I thought overfitting means - producing a model maximizing performance on existing observations but not general enough for unseen data. A more general model would have shown a smaller value premium.
@@michaelspiano2022 Researchers have to make assumptions when examining historical data. I don't think Hasler is arguing his assumptions are necessarily better or more reasonable than FF, but rather they're just as reasonable and often it's uncertain what the most reasonable assumptions are.
Would be great to have time stamps for all videos!
Super interesting! Thanks to Ben, Cam, and Mr Hasler
Great content as always. Your content continues to empower so many people. Thanks to both of you and the whole team that helps get these episodes published :)
1. Hedging against inflation actually equals timing the market and is never a good idea. Just keep to your strategy.
2. Climate should not be part of your investment strategies. Just diversify and keep costs low. Live green, invest smart.
3. Value investing might give more returns over really long terms, but to me it seems not worth the risk. Sometimes it takes a decade or longer for the value factor to show up. What if my horizon of 25 years is too short for the value factor to emerge when tilting to value? Second, there is no good ETF available for me to even try this. I stick to a globally diversified portfolio and with that almost guarantee great results.
Edit: Great video and discussion and congrats on your success on the website and in general. You tought me and some friends a lot of how to invest smart!
Now THAT is a common sense and rational response
This is an important topic since so much of CSI and RR is about the value factor.
Does this change your view on value being a robust factor? And that a small-cap value tilt is still optimal for those willing to take on more complexity and risk?
Fair question, though he doesn't argue the value factor doesn't exist. He doesn't argue his assumptions are necessarily better, rather there are any number of reasonable, but arbitrary assumptions made in financial research. Even if it is smaller, the value premium still seems statistically significant.
I subscribe to Professor Andrew Lo's adaptive views of the markets so though the value factor may have taken a hit for decades, capitalism's mean reverting characteristics should suggest a resurgence of value but irritatingly, the question is when.
With that being said, instead of trying to time and ride the possible wave of value resurgence, I'll just stick to my market cap weighted index funds.
I have light factor slants, but as Ben would say, there are no good arguments against your approach either.
I tried to create an account in the Rational Reminder Community and was rejected because of "Suspicious IP". I live in Brazil, so that's probably why, but I am a legit fan, I have messages in Ben's videos from years ago
Apply again and put this in the note.
@@rationalreminder I was accepted. Thanks!
Right when I was about to finally go to sleep. Looks like I’ll be breaking dawn
just watch it while u sleep g
“Hm.”
Ben is not impressed
There are easier people too impress than Ben :)
But is it justified to take out "climate shocks"? Shouldn't we expect more shocks as the climate gets worse?
If we expect them, they are not shocks.