Chart: The Futility of Stock Market Prediction
- SimpleGift
- Posts: 4477
- Joined: Tue Feb 08, 2011 2:45 pm
Chart: The Futility of Stock Market Prediction
Last month, Vanguard published an updated research report comparing the effectiveness of 16 different metrics for predicting stock market returns 1 and 10 years in advance (chart below). Even the most often-discussed and effective metric (P/E 10, Shiller CAPE) only explained 43% of the variance in stock market returns 10 years ahead. In fact, rainfall scored in 8th place out the 16 metrics — far ahead of corporate profit margins, trend earnings growth and consensus GDP growth!
Source: Vanguard Research
What's truly sad is all the time, effort and resources that the financial media spends in analyzing and discussing these metrics.
Source: Vanguard Research
What's truly sad is all the time, effort and resources that the financial media spends in analyzing and discussing these metrics.
Last edited by SimpleGift on Sun Dec 02, 2012 5:40 pm, edited 1 time in total.
Re: Chart: The Futility of Stock Market Prediction
Two comments:
1) Social scientists often use R^2 in these regions.
2) The absence of a single variable having a variance greater than 0.43 is not proof of its absence.
Beware of herd thinking, i.e. personal confirmational bias echo'd anywhere, including on this forum
1) Social scientists often use R^2 in these regions.
2) The absence of a single variable having a variance greater than 0.43 is not proof of its absence.
Beware of herd thinking, i.e. personal confirmational bias echo'd anywhere, including on this forum
Re: Chart: The Futility of Stock Market Prediction
It's hard to predict rain, but rain has some predictive qualities. Interesting. I'll stay the course according to my IPS.
Jerry
Jerry
"I was born with nothing and I have most of it left."
- Taylor Larimore
- Posts: 32839
- Joined: Tue Feb 27, 2007 7:09 pm
- Location: Miami FL
The Futility of Stock Market Prediction
Bogleheads:
For anyone who believes you can forecast the stock market, I invite you to enter the 2013 Boglehead Contest. Registration will be on this forum during the first ten days of January.
Best wishes.
Taylor
For anyone who believes you can forecast the stock market, I invite you to enter the 2013 Boglehead Contest. Registration will be on this forum during the first ten days of January.
Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
Re: Chart: The Futility of Stock Market Prediction
True, but I assume they wouldn't do it if no one read it.Simplegift wrote:Last month, Vanguard published an updated research report comparing the effectiveness of 16 different metrics for predicting stock market returns 1 and 10 years in advance (chart below). Even the most often-discussed and effective metric (P/E 10, Shiller CAPE) only explained 43% of the variance in stock market returns 10 years ahead. In fact, rainfall scored in 8th place out the 16 metrics — far ahead of corporate profit margins, trend earnings growth and consensus GDP growth!
Source: Vanguard Research
What's truly sad is all the time, effort and resources that the financial media spends in analyzing and discussing these metrics.
"Yes, investing is simple. But it is not easy, for it requires discipline, patience, steadfastness, and that most uncommon of all gifts, common sense." ~Jack Bogle
Re: Chart: The Futility of Stock Market Prediction
This has been discussed elsewhere, but if you use 20-year ahead real earnings, R^2 is somewhere around 2/3 which is quite significant. That is just with one factor. If you had some kind of multi-factor model, maybe it would even be higher.Simplegift wrote:Last month, Vanguard published an updated research report comparing the effectiveness of 16 different metrics for predicting stock market returns 1 and 10 years in advance (chart below). Even the most often-discussed and effective metric (P/E 10, Shiller CAPE) only explained 43% of the variance in stock market returns 10 years ahead. In fact, rainfall scored in 8th place out the 16 metrics — far ahead of corporate profit margins, trend earnings growth and consensus GDP growth!
Source: Vanguard Research
What's truly sad is all the time, effort and resources that the financial media spends in analyzing and discussing these metrics.
Does not sound so futile at all. I think you are drawing a foregone conclusion.
Re: Chart: The Futility of Stock Market Prediction
I will stay the course.
Chaz |
|
“Money is better than poverty, if only for financial reasons." Woody Allen |
|
http://www.bogleheads.org/wiki/index.php/Main_Page
Re: Chart: The Futility of Stock Market Prediction
Unfortunately the paper did not resolve the PE10 vs. PE1 debate that has gone on many times here--those two "work" the best but have offered such different "forecasts" of late!
Regardless, the paper does at least show that price matters to some extent, thought it is certainly not determinative.
Regardless, the paper does at least show that price matters to some extent, thought it is certainly not determinative.
- SimpleGift
- Posts: 4477
- Joined: Tue Feb 08, 2011 2:45 pm
Re: Chart: The Futility of Stock Market Prediction
It's possible. Though, I've previously posted research indicating the effectiveness of the Gordon Equation in predicting 30-year stock market expectations (and described by Bill Bernstein as "the closest thing to a physical law, like gravity or planetary motion, as we will ever encounter in finance"). So I'm open to learning about metrics that research indicates are effective.grayfox wrote:I think you are drawing a foregone conclusion.
In this recent study from Vanguard, it was just so surprising to see that most of the metrics analyzed at such great length in the financial press (corporate profits, operating earnings, GDP growth, etc.) appear to have absolutely no predictive power or influence whatever, even for next year's stock returns. Cheers.
Re: Chart: The Futility of Stock Market Prediction
That Vanguard report is excellent because it debunks most of the bogus forecasts.
And the ones that have any effectiveness only apply to the long term. That's the key to it. Valuations forecast returns 20 or 30 years out. Shorter term, forget about it.
I looked into it once and concluded that to be sure that he gets the expected return, an investor must hold stocks for about the payback period or duration, which is somewhere between P/E and P/D. So that would be somewhere between 21.66 and 47.8 years.
Also, in the really long term, it's unpredictable. As t->infinity, the probability of a black swan event that wipes out everything, approaches 1.
And the ones that have any effectiveness only apply to the long term. That's the key to it. Valuations forecast returns 20 or 30 years out. Shorter term, forget about it.
I looked into it once and concluded that to be sure that he gets the expected return, an investor must hold stocks for about the payback period or duration, which is somewhere between P/E and P/D. So that would be somewhere between 21.66 and 47.8 years.
Also, in the really long term, it's unpredictable. As t->infinity, the probability of a black swan event that wipes out everything, approaches 1.
-
- Posts: 101
- Joined: Tue Feb 22, 2011 8:45 pm
- Location: Portland, OR
- Contact:
Re: Chart: The Futility of Stock Market Prediction
Sometimes I like to sit up late in my office, swirling my drink and stroking a black cat while I tinker with statistics like these. According to my exploratory spreadsheet, the following predictive variables are in the same ballpark as p/e 10 and p/e 1 for a 10 year outlook (using a simple 10 yr moving average of annual Dow returns):
prime rate:
Positive correlation. No surprise here, I've observed that the market appears to digest interest rates on about a six year lag.
Tobin's q ratio:
Negative correlation. Kind of a price/book variant with some built-in compensation for inflation (as I understand it). I'm a bit surprised this wasn't in the study as it's a noted predictive variable.
marginal tax rate:
Negative correlation. Again, seems fairly intuitive.
unemployment rate:
Positive correlation. Seems a bit counter-intuitive. Perhaps a loose labor market favors earnings? Might be worth mentioning here that my dataset is shorter than Vanguard's and doesn't include the Great Depression.
m/o demographic ratio:
Negative correlation. Ratio of sizes of middle age to old age cohorts as detailed at http://www.frbsf.org/publications/econo ... 11-26.html Once you get into these time frames having a smaller middle age cohort and a larger old age cohort seems to bode well (not at all the same case as in the short run). Possibly long-term cycles showing up (inter-generational wealth transfers?).
p/e volatility:
Negative correlation. An original metric perhaps? A cursory examination doesn't find any references to it. To get volatility I calculate each year's variation from the 5 yr trend in p/e 1 and then figure 3 yr rolling absolute averages of those variations). Might serve as a proxy for business "visibility": more prosaic times yield better returns.
prime rate:
Positive correlation. No surprise here, I've observed that the market appears to digest interest rates on about a six year lag.
Tobin's q ratio:
Negative correlation. Kind of a price/book variant with some built-in compensation for inflation (as I understand it). I'm a bit surprised this wasn't in the study as it's a noted predictive variable.
marginal tax rate:
Negative correlation. Again, seems fairly intuitive.
unemployment rate:
Positive correlation. Seems a bit counter-intuitive. Perhaps a loose labor market favors earnings? Might be worth mentioning here that my dataset is shorter than Vanguard's and doesn't include the Great Depression.
m/o demographic ratio:
Negative correlation. Ratio of sizes of middle age to old age cohorts as detailed at http://www.frbsf.org/publications/econo ... 11-26.html Once you get into these time frames having a smaller middle age cohort and a larger old age cohort seems to bode well (not at all the same case as in the short run). Possibly long-term cycles showing up (inter-generational wealth transfers?).
p/e volatility:
Negative correlation. An original metric perhaps? A cursory examination doesn't find any references to it. To get volatility I calculate each year's variation from the 5 yr trend in p/e 1 and then figure 3 yr rolling absolute averages of those variations). Might serve as a proxy for business "visibility": more prosaic times yield better returns.
Re: Chart: The Futility of Stock Market Prediction
This chart (with single parameters) brings to mind this joke:Simplegift wrote:What's truly sad is all the time, effort and resources that the financial media spends in analyzing and discussing these metrics.
If it were not possible to predict stock prices, then algorithmic program traders would not be making money. Obviously, some people have figured out in some time frames and under some defined parameters to make money predicting price movements. Isn't it enough for BH's to be happy to make money the "BH way" and leave it at that? The people who know how, are wise not to let on. To say it is impossible seems to me to be forgetting insider trading, sophisticated algorithms or making pronouncements not far from religious dogma.Late at night, a drunk was on his knees beneath a street-light, evidently looking for something.
A passer-by, being a good Samaritan, offered to help. “What is it you have lost? ” he asked.
”My watch, ” replied the drunk. “It fell off when I tripped over the pavement.”
The passer-by joined in the search but after a quarter of an hour, there was still no sign of the watch.
“Where exactly did you trip? ” asked the passer-by.
“About half a block up the street, ” replied the drunk.
“Then why are you looking for your watch here if you lost it half a block up the street? ”
The drunk said: “Because the light’s a lot better here. ”
source: http://www.jokebuddha.com/joke/Under_th ... z2E2wQe0wk
Re: Chart: The Futility of Stock Market Prediction
The chart actually suggests PE10 and PE1 have predictive power--quite a lot of it, in fact.
Re: Chart: The Futility of Stock Market Prediction
Ah-ha! But rainfall where is the real question. Figure that out and it will all be gravy.
LOSER of the Boglehead Contest 2015 |
lang may yer lum reek
- SimpleGift
- Posts: 4477
- Joined: Tue Feb 08, 2011 2:45 pm
Re: Chart: The Futility of Stock Market Prediction
One question does come to mind though: What practical use can an investor make of a metric that predicts 40% of the variation in stock market returns 10 years in advance?exeunt wrote:The chart actually suggests PE10 and PE1 have predictive power--quite a lot of it, in fact.
Neither PE10 or PE1 appears to provide nearly enough certainty for tactical asset allocation signals (i.e., when to sell stocks and when to buy them back). I can see how these metrics might be helpful in estimating whether 10-year stock returns will be above or below average (higher valuations = lower returns and vice-versa). But even this estimate would seem to involve a large degree of error, as 60% of the variation in returns at 10 years is still unexplained by valuation. Just wondering what the practical use to an investor might be?
Last edited by SimpleGift on Mon Dec 03, 2012 10:47 pm, edited 1 time in total.
Re: Chart: The Futility of Stock Market Prediction
Well, you could decide to bump up your equity allocation by 10% if PE and PE10 both indicated returns were expected to be higher than normal.Simplegift wrote:One question does come to mind though: What practical use can an investor make of a metric that predicts 40% of the variation in stock market returns 10 years in advance?exeunt wrote:The chart actually suggests PE10 and PE1 have predictive power--quite a lot of it, in fact.
Neither PE10 or PE1 appears to provide nearly enough certainty for tactical asset allocation signals (i.e., when to sell stocks and when to buy them back). I can see how these metrics might be helpful in estimating whether 10-year stock returns will be above or below average (higher valuations = lower returns and vice-versa). But even this estimate would seem to involve a large degree of error, as 60% of the variation in returns at 10 years is still unexplained by valuation. Just wondering.
The fact that nobody can easily exploit and arbitrage away an inefficiency on a decade or longer time scale is why one can exist in the first place.
Most of my posts assume no behavioral errors.
Re: Chart: The Futility of Stock Market Prediction
In most cases, it won't. But when valuations get extreme, PE10 can provide a useful signal.Simplegift wrote:One question does come to mind though: What practical use can an investor make of a metric that predicts 40% of the variation in stock market returns 10 years in advance?exeunt wrote:The chart actually suggests PE10 and PE1 have predictive power--quite a lot of it, in fact.
Neither PE10 or PE1 appears to provide nearly enough certainty for tactical asset allocation signals (i.e., when to sell stocks and when to buy them back). I can see how these metrics might be helpful in estimating whether 10-year stock returns will be above or below average (higher valuations = lower returns and vice-versa). But even this estimate would seem to involve a large degree of error, as 60% of the variation in returns at 10 years is still unexplained by valuation. Just wondering what the practical use to an investor might be?
- SimpleGift
- Posts: 4477
- Joined: Tue Feb 08, 2011 2:45 pm
Re: Chart: The Futility of Stock Market Prediction
Thank you, baw and exeunt. You inspired me to take a look at a scatterplot of the actual Shiller PE10 return values (below). While I can see an obvious case for shifting a portion of the portfolio when PE10 gets over 35 (which looks fairly rare), the rest of the data points look a little too widely scattered around the trendline for me to have much conviction in them. With PE10 below 10, however, even with the wide dispersion of returns, this whole range of returns is so positive (not one negative data point and only one under 3% real) that stocks would likely beat bonds in nearly any scenario — and I'd also be tempted to shift a portion of my portfolio. Thanks!baw703916 wrote:Well, you could decide to bump up your equity allocation by 10% if PE and PE10 both indicated returns were expected to be higher than normal.
Source: Money Blog
Re: Chart: The Futility of Stock Market Prediction
Algorithmic trading can make money without predicting future stock prices by cannibalising the old fashioned market maker’s business, performing the same function at a fraction of the cost. It can lower transaction costs for all market participants by tightening spreads and providing liquiditywshang wrote:If it were not possible to predict stock prices, then algorithmic program traders would not be making money. Obviously, some people have figured out in some time frames and under some defined parameters to make money predicting price movements. Isn't it enough for BH's to be happy to make money the "BH way" and leave it at that? The people who know how, are wise not to let on. To say it is impossible seems to me to be forgetting insider trading, sophisticated algorithms or making pronouncements not far from religious dogma.
Re: Chart: The Futility of Stock Market Prediction
While what you write is correct, there are algorithmic traders making money off statistical correlations of historic movements; in other words, market inefficiencies not evident to the masses. Here is an example where I could write more, but it is not in my interest to say any more. Good luck - the BH way is the best way for the great majority of us.Verde wrote:Algorithmic trading can make money without predicting future stock prices by cannibalising the old fashioned market maker’s business, performing the same function at a fraction of the cost. It can lower transaction costs for all market participants by tightening spreads and providing liquiditywshang wrote:If it were not possible to predict stock prices, then algorithmic program traders would not be making money.
The cure shouldn't be worse than the disease.
Re: Chart: The Futility of Stock Market Prediction
It depends on how someone sets their target asset allocation. If your asset allocation takes into account the expected return and expected volatility of assets, changes in valuation may change your target allocation. Or may not change it at all.Simplegift wrote:One question does come to mind though: What practical use can an investor make of a metric that predicts 40% of the variation in stock market returns 10 years in advance?exeunt wrote:The chart actually suggests PE10 and PE1 have predictive power--quite a lot of it, in fact.
For example, suppose at some time you forecast the following long-term real expected return,standard deviations: stocks 7,20 and bonds 2,5. You choose portfolio weights 50/50 and forecast 4.5,12.5 for your portfolio. (ignoring covariance.)
Then stock valuations go way up, and you forecast: stocks 2,20 and bonds 2,5. 50/50 is now forecast 2,12.5 (ignoring covariance). You can stick with 50/50, but you are taking a lot of volatility (12.5%) for only 2% real return. You could reduce volatility by going to 25/75 which would be 2,8.75
This is a simple specific case when stock earnings yield fall to the same as bond yield. Here you can reduce portfolio volatility at the same expected return. In general, the optimum weight may change with changing expectations.
I always see people write, don't do such and such (like invest in individual stocks), because you are taking uncompensated risk. Well if your portfolio has 50% stocks and volatility 12.5%, and you could get the same expected return with 25% stocks and 8.5% volatility, then you are taking uncompensated risk.
Re: Chart: The Futility of Stock Market Prediction
A few years back I made a scatter plot of the 10 year returns going forward vs. the previous 10 year returns, and got a similar looking graph. Kind of makes sense, because valuations only tend to get really high when the market has been on a tear for an extended period.Simplegift wrote:Thank you, baw and exeunt. You inspired me to take a look at a scatterplot of the actual Shiller PE10 return values (below). While I can see an obvious case for shifting a portion of the portfolio when PE10 gets over 35 (which looks fairly rare), the rest of the data points look a little too widely scattered around the trendline for me to have much conviction in them. With PE10 below 10, however, even with the wide dispersion of returns, this whole range of returns is so positive (not one negative data point and only one under 3% real) that stocks would likely beat bonds in nearly any scenario — and I'd also be tempted to shift a portion of my portfolio. Thanks!baw703916 wrote:Well, you could decide to bump up your equity allocation by 10% if PE and PE10 both indicated returns were expected to be higher than normal.
When people hear about the Lost Decade or the Death of Equities, their first reaction usually isn't Buy! Buy! Buy! Nor is the typical reaction fo Dow 36,000 to load up on TIPS and Treasuries.
Most of my posts assume no behavioral errors.
Re: Chart: The Futility of Stock Market Prediction
Outside of maybe big bubbles and big anti-bubbles it is hard to see how to make much of any of this. You would have to make an asset allocation based on today's predictor and hold for 20 year or 30 years. How many people put down a large chunk of change and let it sit that long? Could work if you are say less than 30 years old or so and have lots saved, but most people are only really getting up a head of investing steam by then. Most also adjust allocation for risk along the way (age in bonds or whatever) and so make adjustments more frequently than ever 20-30 years.
Lots of people, including myself, have looked at valuation base switching algorithms. Given you know the data and can fine tune to fit the known data, sure, you can make something that "works". Making this work going forward on data you have not seen is entirely different and I have never seen any evidence this is likely to work to your benefit.
Even in bubbles this has proven to be very hard to make work in practice. Sell too early and lose. Sell too late and lose. Buy back in at the wrong time and lose. This is never easy, though some do manage to do well. So would a few people with a random date switch if you handed out enough random dates.
About the only real value as near as I can tell is to adjust expectations of future returns. Ie when valuations are high (likely because market has been screaming up) don't get trapped into thinking this is going to last and you are going to be rich, and if valuations are very low (likely because markets have been tanking) don't get sucked into despair, this too will change.
Lots of people, including myself, have looked at valuation base switching algorithms. Given you know the data and can fine tune to fit the known data, sure, you can make something that "works". Making this work going forward on data you have not seen is entirely different and I have never seen any evidence this is likely to work to your benefit.
Even in bubbles this has proven to be very hard to make work in practice. Sell too early and lose. Sell too late and lose. Buy back in at the wrong time and lose. This is never easy, though some do manage to do well. So would a few people with a random date switch if you handed out enough random dates.
About the only real value as near as I can tell is to adjust expectations of future returns. Ie when valuations are high (likely because market has been screaming up) don't get trapped into thinking this is going to last and you are going to be rich, and if valuations are very low (likely because markets have been tanking) don't get sucked into despair, this too will change.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
- Clearly_Irrational
- Posts: 3087
- Joined: Thu Oct 13, 2011 3:43 pm
Re: Chart: The Futility of Stock Market Prediction
I've been wondering lately if anyone has done modeling on using the PE10 to adjust your equity share. Say for example your default share is 60% but you ramp it up/down asymptotically in a band that ranges from say 40%-80% depending on the PE10 values .Simplegift wrote:Thank you, baw and exeunt. You inspired me to take a look at a scatterplot of the actual Shiller PE10 return values (below). While I can see an obvious case for shifting a portion of the portfolio when PE10 gets over 35 (which looks fairly rare), the rest of the data points look a little too widely scattered around the trendline for me to have much conviction in them. With PE10 below 10, however, even with the wide dispersion of returns, this whole range of returns is so positive (not one negative data point and only one under 3% real) that stocks would likely beat bonds in nearly any scenario — and I'd also be tempted to shift a portion of my portfolio. Thanks!
- SimpleGift
- Posts: 4477
- Joined: Tue Feb 08, 2011 2:45 pm
Re: Chart: The Futility of Stock Market Prediction
Good question. Just looking at the history of the Shiller PE10 metric (chart below), it appears that if the thresholds are set at just the extremes of valuation (say, PE10 greater than 30 and less than 10) — as was suggested upthread — then you'd only be adjusting your portfolio perhaps once or twice in your lifetime — sort of a rare "John Boglesque" tilt based on extreme valuations. I'd think anything more frequent would be difficult to implement due to confusing signals, as Rodc suggests.Clearly_Irrational wrote:I've been wondering lately if anyone has done modeling on using the PE10 to adjust your equity share. Say for example your default share is 60% but you ramp it up/down asymptotically in a band that ranges from say 40%-80% depending on the PE10 values .
Source: multpl.com
Re: Chart: The Futility of Stock Market Prediction
Yeah, some subjects get brought back almost annually. Here is a discussion about Tactical Asset Allocation based upon PE10. Go the the bottom and you see a table by FredFlinstone answering your question. It is interesting to see almost the same comments by the same people.Clearly_Irrational wrote:I've been wondering lately if anyone has done modeling on using the PE10 to adjust your equity share. Say for example your default share is 60% but you ramp it up/down asymptotically in a band that ranges from say 40%-80% depending on the PE10 values .
http://www.bogleheads.org/forum/viewtop ... &start=150
The cure shouldn't be worse than the disease.
- Taylor Larimore
- Posts: 32839
- Joined: Tue Feb 27, 2007 7:09 pm
- Location: Miami FL
Stay the Course
When I started investing in 1950 the S&P was less than 20. It is now over 1,400 (not counting dividends). I wasted many years trying to time the market when I should have just "stood there."I've been wondering lately if anyone has done modeling on using the PE10 to adjust your equity share. Say for example your default share is 60% but you ramp it up/down asymptotically in a band that ranges from say 40%-80% depending on the PE10 values .
Best wishes.Stay the course. No matter what happens, stick to your program. I've said 'Stay the Course" a thousand times, and I meant it every time. It is the most important single piece of investment wisdom I can give to you. -- Jack Bogle in Common Sense on Mutual Funds
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
- Clearly_Irrational
- Posts: 3087
- Joined: Thu Oct 13, 2011 3:43 pm
Re: Stay the Course
Looking for the magic formula is what lead me to buy and hold indexing in the first place. I find that the more I explore possible strategies, the less likely I am to use most of them since they usually turn out to be so much bunk. That said, momentum and herd mentality are well known investor phenomenons. I made money off that back when I traded individual stocks but eventually decided it couldn't be made into a reliable system. Is that true for the entire market? I'm not sure yet, and I've seen little evidence one way or the other.Taylor Larimore wrote: When I started investing in 1950 the S&P was less than 20. It is now over 1,400 (not counting dividends). I wasted many years trying to time the market when I should have just "stood there."
Re: Chart: The Futility of Stock Market Prediction
Mr. Bogle has been willing to forecast what he expects the return of stocks to be on a decade time scale based on the Gordon equation. So I don't see how one can say that valuations don't matter at all.
Most of my posts assume no behavioral errors.
Re: Chart: The Futility of Stock Market Prediction
What happens is that any time the subject of valuation is brought up, the discussion degenerates into a discussion of market timing using some kind of switching or ramping. This has been discussed to death, and none of it makes any sense and is a waste of time to discuss further.
But it is a straw man. It does not mean that valuations do not matter, or that knowing the expected return is useless.
But it is a straw man. It does not mean that valuations do not matter, or that knowing the expected return is useless.
Re: Chart: The Futility of Stock Market Prediction
I've never heard of a market timing newsletter that came out once every five years, which is pretty much the minimum time scale for valuations to have any predictive ability. Every 15 or 20 years would be better.
Most of my posts assume no behavioral errors.
- SimpleGift
- Posts: 4477
- Joined: Tue Feb 08, 2011 2:45 pm
Re: Chart: The Futility of Stock Market Prediction
In trying to sum up this thread so far, it could perhaps be said that, for the ordinary investor, stock prediction is indeed "futile," but at the same time it can also be "useful."
In other words, even the most effective metrics of PE10 or PE1 don't provide enough certainty for tactical asset allocation shifts (except perhaps at the most extreme valuations, once or twice per lifetime) — but knowing when stocks are overvalued or undervalued can help provide the ordinary investor with a "sanity check" when markets are getting irrationally exuberant or pessimistic.
And, finally, like the Gordon Equation, having a sense of expected returns based on current valuations can help the investor set realistic goals for portfolio construction and retirement planning. At least these are my takeaways from this thread!
In other words, even the most effective metrics of PE10 or PE1 don't provide enough certainty for tactical asset allocation shifts (except perhaps at the most extreme valuations, once or twice per lifetime) — but knowing when stocks are overvalued or undervalued can help provide the ordinary investor with a "sanity check" when markets are getting irrationally exuberant or pessimistic.
And, finally, like the Gordon Equation, having a sense of expected returns based on current valuations can help the investor set realistic goals for portfolio construction and retirement planning. At least these are my takeaways from this thread!