Interesting study on impact of providing investors feedback

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 9:45 am

http://www.etf.com/sections/index-investor-corner/swedroe-investors-respond-feedback

The results are quite interesting I thought. Hope you find it helpful
Larry

ResearchMed
Posts: 4720
Joined: Fri Dec 26, 2008 11:25 pm

Re: Interesting study on impact of providing investors feedback

Post by ResearchMed » Fri Feb 17, 2017 9:50 am

larryswedroe wrote:http://www.etf.com/sections/index-investor-corner/swedroe-investors-respond-feedback

The results are quite interesting I thought. Hope you find it helpful
Larry


Interesting article/study.

But what does "When this study was controlled for attrition, the authors were unable to explain their results" mean?

RM
This signature is a placebo. You are in the control group.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 9:57 am

Interesting.

A couple questions:

1) The sample reported higher "risk adjusted returns" But what about absolute returns?

2) What does this statement mean? "When this study was controlled for attrition, the authors were unable to explain their results."
edit: Looks like RM beat me asking question 2).
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 10:03 am

re attrition, this from the paper should answer your question
In the literature on learning Amit Seru, Tyler Shumway, and Noah Stoffman (2010) show that the learning of private investors in investment matters is mainly driven by attrition. Investors who learn about their inferior investment skills stop trading. Against this background, it is important to check whether the results we presented above are also driven by attrition. In our sample, all investors from the group of subscribers as well as from the group of nonsubscribers are equally present and active throughout the entire sample period. However, in the banking industry it is well documented that investors do not close their accounts but rather stop trading and/or reduce their account to small balances (Stephanie Coyles and Timothy C. Gokey 2002; Rex Y. Du, Wagner A. Kamakura, and Carl F. Mela 2007). This effect is usually known as “silent-attrition”.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 10:49 am

larryswedroe wrote:re attrition, this from the paper should answer your question
...

To rephrase my question 1) Re: improved risk adjusted return metric.

You taught us about need/willingness/ability. Who cares about risk adjusted returns when one is 85 and not working? One cares about dollars.

I'm not saying trading / turnover will lead to higher returns (either absolute or risk adjusted). But to neglect absolute returns in the study/article is cherry picking.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 11:11 am

EHENGINEER
Sorry IMO have it exactly backwards. One should care about the risk adjusted returns, meaning they adjust returns for the exposure to factors, like stocks and value and small. So you don't get credit for higher returns simply because you took more risk.

Larry

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 11:14 am

larryswedroe wrote:EHENGINEER
Sorry IMO have it exactly backwards. One should care about the risk adjusted returns, meaning they adjust returns for the exposure to factors, like stocks and value and small. So you don't get credit for higher returns simply because you took more risk.

Larry


Larry, False dichotomy. You should study both. You can't eat your sharpe ratio in retirement.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

User avatar
SeeMoe
Posts: 852
Joined: Sat Jul 18, 2015 11:30 am
Location: Near Philly..

Re: Interesting study on impact of providing investors feedback

Post by SeeMoe » Fri Feb 17, 2017 11:15 am

I saw myself in that article over and over throughout my investing years. I finally had a wake up call , for a small fee that paid for itself many, many times over, after 18 years of DIY when I had a 1x folio analysis done by Vanguard Services. My crystal ball investing technique was shattered,..hopefully forever, and My new folio has make us Flagship heros from going nowhere zeros!

SeeMoe.. :mrgreen:
"By gnawing through a dike, even a Rat can destroy a nation ." {Edmund Burke}

ResearchMed
Posts: 4720
Joined: Fri Dec 26, 2008 11:25 pm

Re: Interesting study on impact of providing investors feedback

Post by ResearchMed » Fri Feb 17, 2017 11:21 am

larryswedroe wrote:EHENGINEER
Sorry IMO have it exactly backwards. One should care about the risk adjusted returns, meaning they adjust returns for the exposure to factors, like stocks and value and small. So you don't get credit for higher returns simply because you took more risk.

Larry


Taking this to an extreme would suggest something extremely cautious, "non risky", and very low returns.

Not at all what one (or should I write, "the two of us") wants for very long term planning.

We'll take some "more risk".
That's been our strategy for a long time, and we've done well.

("Too risky" is in the eye of the beholder, etc., and also often pertains to time frame, plus the regular "need/ability" to take risk, etc.)

RM
This signature is a placebo. You are in the control group.

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 11:33 am

EH
Of course you cannot eat your Sharpe ratio, but we aren't discussing that. IT's just how you did perform against an APPROPRIATE benchmark. Otherwise it's like arguing you did well by purchasing junk bonds and comparing returns to Tbills.
Larry

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 11:35 am

ResearchMed
I have NO idea how you draw that conclusion. It doesn't apply in any way to what I wrote. What we are simply talking about is measuring things properly.
Larry

ResearchMed
Posts: 4720
Joined: Fri Dec 26, 2008 11:25 pm

Re: Interesting study on impact of providing investors feedback

Post by ResearchMed » Fri Feb 17, 2017 11:40 am

larryswedroe wrote:ResearchMed
I have NO idea how you draw that conclusion. It doesn't apply in any way to what I wrote. What we are simply talking about is measuring things properly.
Larry


Your: "So you don't get credit for higher returns simply because you took more risk.

But you are technically correct anyway.
It isn't "credit" that we care about.

It's those "higher returns" due to being willing and able to take some "more risk".

RM
This signature is a placebo. You are in the control group.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 12:05 pm

larryswedroe wrote:EH
Of course you cannot eat your Sharpe ratio, but we aren't discussing that. IT's just how you did perform against an APPROPRIATE benchmark. Otherwise it's like arguing you did well by purchasing junk bonds and comparing returns to Tbills.
Larry


The fact that you/authors choose to ignore total return metric hints at cherry picking. The paper/article is talking about influencing investor behavior. Positive affects were discussed, but neglected potential negative affects - ie lower total returns. I assert that all effects should be studied.

Maybe another analogy will help?

"We manipulated a bunch of investors into selling all stock and buying short term treasuries. Investors improved Sharpe Ratios!!!" Analogous to study. And obviously should be noted that lower sharpe ratios may be beneficial and improve probability of higher returns. This is your own need/willingness/ability story and should not be neglected in studies of influencing investor behavior.
Last edited by EHEngineer on Fri Feb 17, 2017 12:16 pm, edited 1 time in total.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 12:14 pm

ResearchMed wrote:Your: "So you don't get credit for higher returns simply because you took more risk.

But you are technically correct anyway.
It isn't "credit" that we care about.

It's those "higher returns" due to being willing and able to take some "more risk".

RM

Yes.

Typically Larry like to speak about "return for a given level of risk" But in my exprience people think "Is it worth 5% standard deviation for 2% expected return?" (even if they don't understand they think that way).

So less relevant the "most efficient point," and more relevant to ask "what investments have an acceptable risk/return tradeoff?" And then they start studying individual stock to find the "best" ones. And maybe they are right. That's the empirical question that Larry/authors failed to answer in this study.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 12:15 pm

Research Med and EH
Both of you again are missing the point. Sure earning a higher return is good, them but it doesn't measure SKILL if not adjusted for risk. You'd all be complaining if a fund manager claimed alpha when beating the S&P 500 but did so by buying small value stocks. This is EXACTLY the same thing. If you don't see that nothing more I can say

Larry
Last edited by larryswedroe on Fri Feb 17, 2017 12:17 pm, edited 1 time in total.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 12:16 pm

larryswedroe wrote:Research Med
you again are missing the point. Sure you earn them but it doesn't measure SKILL which is what the study is measuring.
Larry


Define "SKILL"
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 12:19 pm

You measure or attempt to measure skill by whether you beat a RISK ADJUSTED appropriate benchmark. Now might beat it by random luck. That is what t-stats help us do. But you don't measure skill by comparing oranges with apples.
Seriously, I cannot believe I'm having this conversation
And EH what they did was EXACTLY to find the answer to the question you said they didn't.
Nothing more can add here
Larry

livesoft
Posts: 55514
Joined: Thu Mar 01, 2007 8:00 pm

Re: Interesting study on impact of providing investors feedback

Post by livesoft » Fri Feb 17, 2017 12:25 pm

Perhaps a take-home message from this article is that a place like Vanguard should have their client login splash-screen organized like one of these reports to discourage any client transactions and thus cost Vanguard less in costs related to clients.

I did read the paper linked in Larry's blog article which did show examples of the four types of monthly reports which were rather simple. It seems the average account size was under 100,000 euros and I didn't get the sense that the clients had a particularly long history of owning such accounts.

It might be interesting to see a study that plots of "Account Age" versus "Average Number of Transactions in a Month" or better versus "Average number of sell transactions in a Month" in order to see if individual investors get into buy-and-hold-and-rebalance and/or stay-the-course just based on experience. You know, the excitement of setting up a portfolio and trading diminishes with time and habituation along with perhaps the realization that one isn't going to get rich trading and will lose money. That's the attrition part.
This signature message sponsored by sscritic: Learn to fish.

ResearchMed
Posts: 4720
Joined: Fri Dec 26, 2008 11:25 pm

Re: Interesting study on impact of providing investors feedback

Post by ResearchMed » Fri Feb 17, 2017 12:30 pm

EHEngineer wrote:
ResearchMed wrote:Your: "So you don't get credit for higher returns simply because you took more risk.

But you are technically correct anyway.
It isn't "credit" that we care about.

It's those "higher returns" due to being willing and able to take some "more risk".

RM

Yes.

Typically Larry like to speak about "return for a given level of risk" But in my exprience people think "Is it worth 5% standard deviation for 2% expected return?" (even if they don't understand they think that way).

So less relevant the "most efficient point," and more relevant to ask "what investments have an acceptable risk/return tradeoff?" And then they start studying individual stock to find the "best" ones. And maybe they are right. That's the empirical question that Larry/authors failed to answer in this study.


I think you've captured it nicely where you point out that it isn't necessarily the "most efficient point".

Some investors will have little tolerance for risk, and for them, the "right" choice (or "the right choices") may be quite different vs those who do have the tolerance for risk, along with the "likely" (not guaranteed, of course) higher return.

Otherwise, similar to what you pointed out, we "should" all buy those short term treasuries (or similar).
At the other extreme, perhaps, are those who invest some money in IPO's, being willing (and - hopefully! - able) to take that much higher risk, in hopes of a possibly much higher return.

There are obviously different needs and preferences wrt risk/return, full stop.

Added: I'm not sure this is a "skill" contest in the way that Larry seems to be putting it.
We each have different needs/desires for investing.
Presumably, the right "skill" would be affected by the desired goals.

(If we all had the exact same preferences, I'm not sure just what would happen to "market pricing", but that's a different question.)

RM
This signature is a placebo. You are in the control group.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 12:31 pm

larryswedroe wrote:You measure or attempt to measure skill by whether you beat a RISK ADJUSTED appropriate benchmark. Now might beat it by random luck. That is what t-stats help us do. But you don't measure skill by comparing oranges with apples.
Seriously, I cannot believe I'm having this conversation
And EH what they did was EXACTLY to find the answer to the question you said they didn't.
Nothing more can add here
Larry

I have had this conversation probably 1000 times with stock pickers. Now I am playing devil's advocate and you are failing the same way I fail in these conversations.

Stock pickers ignore "risk adjusted." Total returns rule. Risk adjusted returns are like "participation awards" for 1st graders.

"Larry, our stocks trounced your portfolio and now we're rich. But you tried hard and you earned the Sharpe Ratio Award!!!"


In your terms, they are expressing high willingness and ability. Presumaby also need. And if they come out ahead, you cannot say that the manipulation was good for them. Would like to know the answer that question. Based on study data: was total return higher or lower for those who reduced trading?

It's possible the manipulation provided by the study resulted in lower returns for those affected. If so, is this a violation of fiduciary responsibility for the 401k administrator?
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

pkcrafter
Posts: 11741
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Interesting study on impact of providing investors feedback

Post by pkcrafter » Fri Feb 17, 2017 12:59 pm

The only thing I got from what the authors said about attrition--it has been a factor in previous tests, i.e. participants stopped trading all together--was there was no appreciable attrition in their study and so it was not a factor. There also was no difference considering the four different reports that were presented to participants as the authors expected. All the same, performance and habits improved after reports were provided.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Fri Feb 17, 2017 1:01 pm

pkcrafter wrote:The only thing I got from what the authors said about attrition--it has been a factor in previous tests, i.e. participants stopped trading all together--was there was no appreciable attrition in their study and so it was not a factor. There also was no difference considering the four different reports that were presented to participants as the authors expected. All the same, performance and habits improved after reports were provided.

Paul

Define "performance"
I assert that total return performance was neglected.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Fri Feb 17, 2017 1:49 pm

EH


Stock pickers ignore "risk adjusted." Total returns rule.

That's an example, like buying junk bonds and outperforming treasuries, it has nothing to do with picking individual securities. It's about asset allocation---now if you beat a junk bond index that is what matters. Same with picking stocks. That's why literally all the studies done in finance now use risk-adjusted measures

And in has nothing to do as you suggest with ability, willingness or need to take risk, but the sources of risk and return.

With that said, nothing more I can say
Larry

pkcrafter
Posts: 11741
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Interesting study on impact of providing investors feedback

Post by pkcrafter » Fri Feb 17, 2017 2:21 pm

EHEngineer wrote:
pkcrafter wrote:The only thing I got from what the authors said about attrition--it has been a factor in previous tests, i.e. participants stopped trading all together--was there was no appreciable attrition in their study and so it was not a factor. There also was no difference considering the four different reports that were presented to participants as the authors expected. All the same, performance and habits improved after reports were provided.

Paul

Define "performance"
I assert that total return performance was neglected.


Total return was not mentioned, but improved behavior was shown. I do agree with you that the paper has some flaws.

Overall, we find that our reports help investors improve on trading costs, diversification and performance.

William N. Goetzmann and Alok Kumar (2008) argue that underdiversification is associated with a return loss of 2%. In relation to overtrading, the literature argues that private investors are overconfident, and buy and sell stocks too frequently and consequently do not benefit from their trading (Odean 1998, 1999; Barber and Odean 2001). Brad M. Barber and Terrance Odean (2000) report a return loss of 5% for net returns.


Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

betablocker
Posts: 235
Joined: Mon Jan 11, 2016 1:26 pm

Re: Interesting study on impact of providing investors feedback

Post by betablocker » Fri Feb 17, 2017 5:41 pm

ResearchMed wrote:
larryswedroe wrote:ResearchMed
I have NO idea how you draw that conclusion. It doesn't apply in any way to what I wrote. What we are simply talking about is measuring things properly.
Larry


Your: "So you don't get credit for higher returns simply because you took more risk.

But you are technically correct anyway.
It isn't "credit" that we care about.

It's those "higher returns" due to being willing and able to take some "more risk".

RM


Go ahead and take all the risk you want but you aren't getting the point. If you go to a roulette wheel and happen to win a lot great you have great absolute returns. I don't think I'll invest my money with you based on the results though.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Tue Feb 21, 2017 9:34 am

betablocker wrote:
ResearchMed wrote:
larryswedroe wrote:ResearchMed
I have NO idea how you draw that conclusion. It doesn't apply in any way to what I wrote. What we are simply talking about is measuring things properly.
Larry


Your: "So you don't get credit for higher returns simply because you took more risk.

But you are technically correct anyway.
It isn't "credit" that we care about.

It's those "higher returns" due to being willing and able to take some "more risk".

RM


Go ahead and take all the risk you want but you aren't getting the point. If you go to a roulette wheel and happen to win a lot great you have great absolute returns. I don't think I'll invest my money with you based on the results though.


I get that. I well understand that there is a difference between strategy and outcome. Sometimes people get lucky, and we should not plan on that for future returns.

But this is a research paper studying behavioral techniques to change people's investing habits. And if they want to assert luck-based or any other outcome they should use their data, not hypothetical assumptions.

I assert that when you do a behaviorial study you need to report behaviorial outcomes. Reseachers need to report if people are better off or worse off based on the impact of the study. And they need to do it in the metric that matters in real life: Dollars. Researcher's should not make the hypothetical leap that if the study subjects trade less, they necessarily are better off. They should also not make the hypothetical leap that traders who reduced trading turned into "Risk Factor Aficionados" who can understand and implement strategies based on graduate level statistics. That is the theoretical best case outcome that reported "Improved Sharpe Ratio" implies.

IMO the study result for affected subjects is holding short-term-treasury or some other cash-like investment with exceedingly low returns. It is plausible that this reduced subjects' actual returns, and therefore harmed them in the real world metric that matters: Dollars.

As it stands, we do not know. Study ignored a relevant metric: actual returns.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

User avatar
nisiprius
Advisory Board
Posts: 33641
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Interesting study on impact of providing investors feedback

Post by nisiprius » Tue Feb 21, 2017 9:53 am

In theory, you can trade off a Sharpe ratio any way you like. If you have a higher Sharpe ratio, you should be able to take that in the form of increased returns for the same risk. I think this is exactly what Larry does in "the Larry portfolio," for example.

Comparing Sharpe ratios is better than comparing return, but really it's a muddle when you are comparing situations with different amounts of risk.

If the investors with higher Sharpe ratios are exceeding their risk tolerance, then they are not superior investors even if they have both higher returns and higher Sharpe ratios.

If the investors with lower Sharpe ratios are being unknowingly hypercautious, then they are arguably suboptimal investors, although they may still be called "successful investors" if they are getting sufficient return.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Tue Feb 21, 2017 10:34 am

nisiprius wrote:In theory, you can trade off a Sharpe ratio any way you like. If you have a higher Sharpe ratio, you should be able to take that in the form of increased returns for the same risk. I think this is exactly what Larry does in "the Larry portfolio," for example.

Comparing Sharpe ratios is better than comparing return, but really it's a muddle when you are comparing situations with different amounts of risk.

If the investors with higher Sharpe ratios are exceeding their risk tolerance, then they are not superior investors even if they have both higher returns and higher Sharpe ratios.

If the investors with lower Sharpe ratios are being unknowingly hypercautious, then they are arguably suboptimal investors, although they may still be called "successful investors" if they are getting sufficient return.



Nisi, if you don't understand my concern then maybe I am crazy. But let me try again with some totally made-up example numbers.

Before study
Excess returns 7%
Standard deviation 25%
Sharpe Ratio: 0.28

After study
Excess returns 2%
Standard deviation 4%
Sharpe Ratio: 0.5

Conclusions
Sharpe Ratio Improved.
Actual returns far worse.

If a study affects it's participants in this way, I submit that they are harmed, not helped.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Tue Feb 21, 2017 11:26 am

EH
You really are missing the point entirely.
What the study is showing is skill. So if in your example the stock picking by investors gave them higher actual returns but smaller alphas and lower SRs then the answer is they should have stopped picking individual stocks and just bought passive funds and they would have gotten superior results for the same risks. That is what the study is showing. That is why it's important to adjust for exposures to factors (risks). And that is why ALL studies, not just this one, are done that way now.
Larry

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Tue Feb 21, 2017 12:29 pm

larryswedroe wrote:EH
You really are missing the point entirely.
What the study is showing is skill. So if in your example the stock picking by investors gave them higher actual returns but smaller alphas and lower SRs then the answer is they should have stopped picking individual stocks and just bought passive funds and they would have gotten superior results for the same risks. That is what the study is showing. That is why it's important to adjust for exposures to factors (risks). And that is why ALL studies, not just this one, are done that way now.
Larry


Larry, The title of the study was "Does feedback on personal investment success help?" Which is why I want to know about actual returns. Were study subjects helped or harmed?

Here's your logical leap that causes my issues: "they should have stopped picking individual stocks and just bought passive funds and they would have gotten superior results for the same risks" Of course! Virtually all bogleheads believe that! But the study does not test whether subjects did that! (as far as I can tell). Your assumption is tantamount to saying "We taught people that they eat too much candy and are overweight. They are eating less candy and so we assume that they are eating a well-rounded and healthy diet." that's optimistic, but unfounded (and unlikely, IMO).

Unrelated, but since I have your attention: I have learned a ton from your books, especially Winning Bonds Strategy.

Maybe the absolute return data is in this chart, but I never learned "Difference in Difference" analysis, and don't understand this data.
Image
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Tue Feb 21, 2017 2:10 pm

EH
Again you are simply misinterpreting the results, they show they did a better job of managing the risk-adjusted returns when informed. That means that prior to that they were better off if they wanted more risk to buy passive vehicles and not pick stocks. Afterwards it was less the case. So they benefited. If you don't adjust for risk you are not measuring anything correctly. Likely they would continue to benefit from learning and eventually stock picking stocks
Larry

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Interesting study on impact of providing investors feedback

Post by larryswedroe » Tue Feb 21, 2017 5:14 pm

EH
Just thought worth adding that one key finding was that investors traded less when informed of results. We know from other studies the more investors traded the worse their returns, with investors who trade the most underperforming market by about 10% a year. So clearly that almost certainly improved returns, let alone risk adjusted returns

I would add that the study also found they diversified more, clearly also a good thing. Nowhere did they show they had less equity risk. So the returns were very likely higher even before adjusting for the risks.

IMO you are looking for "boogie man" here trying to find some fault in the study when there isn't any boogie man. And the study was done correctly

Larry

User avatar
Taylor Larimore
Advisory Board
Posts: 25782
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

"Investors Respond to Feedback"

Post by Taylor Larimore » Tue Feb 21, 2017 7:23 pm

Bogleheads:

Larry Swedroe has written an interesting article (where does he get the time) for ETF.com. If you are a trader, or thinking about it, Larry's article is for you:

Investors Respond to Feedback

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

nodenuff2
Posts: 303
Joined: Tue Mar 06, 2012 4:08 pm

Re: "Investors Respond to Feedback"

Post by nodenuff2 » Tue Feb 21, 2017 7:36 pm

Thank you Mr. Larimore for taking so much time to help us little guys. You are a blessing to us.
2014 No. 42 2015 No.342 2016 No. 6 what do I know? "Good bless America land that I love..."

gkaplan
Posts: 6940
Joined: Sat Mar 03, 2007 8:34 pm
Location: Portland, Oregon

Re: "Investors Respond to Feedback"

Post by gkaplan » Tue Feb 21, 2017 8:10 pm

Gordon

User avatar
LadyGeek
Site Admin
Posts: 39141
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Interesting study on impact of providing investors feedback

Post by LadyGeek » Tue Feb 21, 2017 8:33 pm

^^^ I moved Taylor Larimore's thread into here.
To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.

EHEngineer
Posts: 485
Joined: Sat Feb 28, 2015 4:35 pm

Re: Interesting study on impact of providing investors feedback

Post by EHEngineer » Tue Feb 21, 2017 10:42 pm

larryswedroe wrote:EH
Just thought worth adding that one key finding was that investors traded less when informed of results. We know from other studies the more investors traded the worse their returns, with investors who trade the most underperforming market by about 10% a year. So clearly that almost certainly improved returns, let alone risk adjusted returns

I would add that the study also found they diversified more, clearly also a good thing. Nowhere did they show they had less equity risk. So the returns were very likely higher even before adjusting for the risks.

IMO you are looking for "boogie man" here trying to find some fault in the study when there isn't any boogie man. And the study was done correctly

Larry


I understand. But why stack assumptions and force readers to deduce, when they could just analyze & present the data directly? The neglect/avoidance of such a key metric as returns seems crazy to me. that's the whole point of investing after all. I'm not qualified to judge if the study was done correctly, but I remain a skeptic. It's just too obvious of a metric to leave unreported. Or, as I mentioned above, maybe it is presented and I don't understand the statistics.

Regardless, I do appreciate your time and effort. Thanks Larry.
Or, you can ... decline to let me, a stranger on the Internet, egg you on to an exercise in time-wasting, and you could say "I'm probably OK and I don't care about it that much." -Nisiprius

Tanelorn
Posts: 1236
Joined: Thu May 01, 2014 9:35 pm

Re: Interesting study on impact of providing investors feedback

Post by Tanelorn » Wed Feb 22, 2017 2:41 am

EHEngineer wrote:
nisiprius wrote:In theory, you can trade off a Sharpe ratio any way you like. If you have a higher Sharpe ratio, you should be able to take that in the form of increased returns for the same risk. I think this is exactly what Larry does in "the Larry portfolio," for example.

Nisi, if you don't understand my concern then maybe I am crazy. But let me try again with some totally made-up example numbers.

Before study
Excess returns 7%
Standard deviation 25%
Sharpe Ratio: 0.28

After study
Excess returns 2%
Standard deviation 4%
Sharpe Ratio: 0.5

Conclusions
Sharpe Ratio Improved.
Actual returns far worse.

The way you trade off ("eat") your increased Sharpe Ratio is by higher leverage. In the above example, the After person could use 3.5x leverage to get the same 7% return but with only 14% SD, a 40% reduction in risk. Alternatively, they could use 6x leverage to get 12% excess returns, 5% more than before, with a similar level of risk (24% SD vs 25% SD).

How practical this is depends on the absolute level of interest rates, the individual's ability to borrow close to risk free, and most importantly, their confidence that the Sharpe ratio really is that high. You don't want to borrow money to play roulette just because you had a string of lucky nights. But if you're Ed Thorp with a shoe computer that predicts where the ball will land, yeah, borrowing is probably fine.

Post Reply