The Perils of Forecasting

In today’s WSJ Econoblog, there is a terrific discussion on a favorite subject of mine: The Perils of Forecasting.

I’m not saying that just because I am linked in it; this is a subject near and dear to me. Recall last year, I wrote a column for the Apprenticed Investor series titled, The Folly of Forecasting. It discussed the perils of predictions — I find the alliterative titles irresistible — and noted that investors all too often give "predictions" far more weight than they should.

The debate is ably handled between two econ profs: James D. Hamilton, of UC San Diego, who’s blog is Econbrowser, and Kash Mansori, of Colby College, who runs the Angry Bear.

Here’s an excerpt:

JH:  "There is both theoretical and empirical evidence against this martingale hypothesis. When investors care about avoiding risk, for example, there could be higher expected returns in times when the stock is more risky to hold. But there is no question that the martingale description is a pretty good first approximation to the behavior of most stock prices.

Actually, economic theory suggests that not just stock prices but a great number of other economic magnitudes of interest should exhibit near-martingale behavior. Predictable price changes in any commodity that can be stored, for example, can be arbitraged away by increasing or decreasing current inventories. Such inventory adjustment again would result in the current price jumping to the value that best reflects the expected future price. Although costs of storage, changing inventories, and other benefits from holding inventories can introduce the possibility of more complicated dynamics, a martingale proves to be a good approximation to exchange rates and many commodity prices."

Next up: Kash:

KM:  Why do economists sometimes get their forecasts wrong in such systematic and persistent ways? Why aren’t they better at adjusting their forecasts?

I think that such episodes point out that the economy is constantly evolving in subtle but substantial ways that can take a long, long time for economists to understand. The economy of the 1990s was truly different from the economy of the 1980s, in ways that we’re still trying to figure out. The labor market of the past few years is truly different from previous economic recoveries, and we’re not quite sure why.

These continual changes in how the economy works put economists at a huge disadvantage in making forecasts, particularly compared to, for example, meteorologists. The nature of physical and chemical interactions in the atmosphere never change, but the nature of economic interactions change constantly!

Its good stuff — worth the read!


The Perils of Forecasting
ECONOBLOG,January 26, 2006

Public Link (no sub req’d)

What's been said:

Discussions found on the web:
  1. royce commented on Jan 26

    To be fair, Barry, isn’t relying on forecasting inherent in what you do? Seems like the only people involved in trading or investing who aren’t forecasting the future are the people who buy a bunch of index funds and never look at them again (and even they are making a prediction on the future direction of the market- otherwise they’d just buy bonds). Or did I miss an important distinction someplace?

  2. PC commented on Jan 26

    Royce, you wrote:

    “Seems like the only people involved in trading or investing who aren’t forecasting the future are the people who buy a bunch of index funds and never look at them again….”

    Actually traders don’t forecast markets (at least the profitable ones). Traders follow trends. The definition of trend following is you “wait” for a trend to shift and then you “follow” it. No where does it mention the word prediction or forecasting.

  3. PC commented on Jan 26

    Royce, you wrote:

    “Seems like the only people involved in trading or investing who aren’t forecasting the future are the people who buy a bunch of index funds and never look at them again….”

    Actually traders don’t forecast markets (at least the profitable ones don’t). Traders follow trends. The definition of trend following is you “wait” for a trend to shift and then you “follow” it. No where does it mention the word prediction or forecasting.

  4. Jason Ruspini commented on Jan 26

    Royce, yes, I think Barry is forecasting – otherwise he would be a pure “i know nothing” trader, i.e. a technical / trend following / momentum / feedback / herd trader.

    Critics of prediction tend to engage in a lot of hyperbole. I think their target is really overconfidence in forecasting (accuracy) and worship of the “expert”.

  5. old ari commented on Jan 26

    Surely, “Alliterative titles are attractive,( or addictive)”.

  6. B commented on Jan 26

    What would everyone do if there were no one predicting? How would we justify these gigantous staffs of paper pushing economists? Who would make all of the pretty charts you post on this board? I think Jason makes a good point. It’s the overconfidence we place in the supposed experts.

    Frankly I can do an admirable job of forecasting using platinum/gold ratios, copper, gold, equities and bonds. Of course, Steve Forbes has pretty much made a similar statement as has John Murphy and quite a few other notable personalities. That is why trend followers didn’t get creamed during the bubble while the majority of economists kept telling people to remain fully invested in equities.

    Jason, trend following isn’st such a bad thing if you are good at it. It made people billions in oil, copper, gold and platinum over the last few years. It is also what allows one to potentially think oil might be topping given the recent upside down crack spread….implying there is too much emotion driving oil. Don’t need to catch the bottom and don’t need to catch the top. Just need to capture 80% of the trend. You’ll retire living large.

  7. Jason Ruspini commented on Jan 26

    B, I wasn’t necessarily disparaging trend-following. There are different ways to make money. Without fundamental views though, you’re more likely to get shaken-out at bad times when trying to follow trends, and catch, say, -20% instead of 80% of the move.

  8. Lymond commented on Jan 26


    “Alliterative appellations…”

  9. B commented on Jan 26

    Any trend trader worth his salt would never take a 20% draw down. If they do, they aren’t trend traders. And even though I’ve shorted the market recently, it’s based on micro trends as no truly tradable trend has developed to tell anyone we are going down. Same for real estate. Both look toppy but a trend hasn’t developed.

    Markets go up and down for many reasons. I’m sort of curious how one can buy at equity market bottoms when fundamentals are awful? Or, even real estate? Or oil? Fundamentalists usually have a rigid view of data. Their own view. The bear:debt is killing America and the markets are going down. Or, teh bull: economic data suggests we are still doing well so the market should go up. To Barry’s point, neither truly knows. It is guessing based on one’s predisposition based on their psychology. But, the trend trader will take either trade because they have no predisposition. Market wants to go down? Let’s hop on board for a short. Wants to go up? Let’s go long and ride the wave.

    How could one have shorted or stepped aside in 2000, 2001 and 2002 if one invested on fundamentals? They were good. Btw, the returns achieved by shorting during those periods were better than any bull market. The economy had a short recession near the bubble top yet the market cratered continuously. What did fundamentals tell anyone about the 87 crash? Or the 29 crash? Trend traders took advantage of every single situation. The markets alerted trend traders in every single situation. It is a misunderstanding that trend traders are herd traders. Usually, they are against the herd with the formation of a new trend. Hence they are usually contrarian traders. Bw, the open cry craders on the pit floors are trend traders. You know, the ones making $20 million a year? I like to do a little predicting but it is usually based on psychology. I don’t invest using it but……’s fun. Mutual funds on the other hand, are usually not trend traders. They are required to take big bets rather than dovetail into a developing trade for most of their money. That is why they can never beat the markets on average. It’s flipping a coin.

    I’m not razzing fundamental analysis but the biggest bull markets usually start when fundamentals are awful….but new trends start. No one believes the trend because the fundamentals are so wrong. so theyy stand aside. The informed trend trader takes the trade and usually gets onboard well before any fundamentalist jumps on board. In fact, we sometimes see a reversal of trend by the time the fundamentalist, or herd, jumps on.

    I’d highly recommend Covel’s book if you are interested. Barry has it posted on the RHS of the blog page.

  10. Jason Ruspini commented on Jan 26

    I guess the conciliatory tone of my last comment failed. Instead of bickering over your specific points, or picking on individual managers (who i’m somewhat sure were actually featured glowingly in the Covel book — right before the kind of drawdown you mention!), I will just say this: the easier a methodology is to implement and the less skill it requires, the more likely it is that any profits will be competed away.

    This doesn’t mean that there aren’t exceptions which can last for quite some time. Look, this should not be a rehash of ye olde “fundamentalists vs technicians” argument.. both have their place. There is no grand decision to be made between the two.

    The moral of this thread is perhaps “know thyself.” Be honest about your strengths and weaknesses, trade accordingly, and be flexible.

  11. B commented on Jan 27

    No need to bicker. We disagree. There’s room for many opinions because that is all they are. If you can make money doing what you are doing, that’s great.

    But, your commentary about trend trading being easy or less likely to profit because it is less skilled are highly erroneous. It’s also erroneous to believe trend traders think things go on forever. Indeed, the most successful trend trader plays reversals as their entry point. Dovetailing in more and more leverage as the trend develops.

    Psychology is the constant struggle to overachieve in trading. You must have an iron gut to buy reversals when the world appears it is coming to an end and every hard wired part of the amygdala is telling you to be very afraid. That part of our brain serves us well in staying alive but is our biggest detractor to market success. That risk, or fear of it, is the reason why most invesors are failures. It’s easy to talk about it. Very, very hard to actually trade it. That is why the blow hards on CNBC can say whatever they want and the smart money doesn’t listen. They collect a pay check at a day job. Not trading.

    Btw, there are no “fund managers” listed in Covel’s book. Have you actually read it? Doesn’t matter if you haven’t but if you are basing your argument on something you haven’t ready…………….But, Soros is in the book. He’s a trend trader. He’s only the most succesful trader of all time.

  12. Jason Ruspini commented on Jan 27

    Ok, this is devolving into semantics. It just seems that your definition of “trend trader” is more broad than mine, and your definition of “fund manager” much more narrow!

    Successful traders by definition capture a trend on some time frame, but labelling them as “trend traders” doesn’t explain anything. Likewise, there may not be “fund managers” in Covel, but there are certainly people who manage funds in there.

  13. Jason Ruspini commented on Jan 27

    “By definition” was too strong, but the points stand.

Posted Under