It looks like you're new here. If you want to get involved, click one of these buttons!
Article provides a thorough look at the implications and impact of MAXDD & DD levels on Investor's decision making.More than anything else, investing is about managing downside risk
© 2015 Mutual Fund Observer. All rights reserved.
© 2015 Mutual Fund Observer. All rights reserved. Powered by Vanilla
Comments
How long did it really take to recover, considering deflation (in the 30s) and dividends? Mark Hulbert wrote this article in the NYTimes, entitled: "25 Years to Bounce Back? Try 4½"
Even using raw stock prices, that's 25 years for the Dow (Nov 23, 1954) per Hulbert, or about 30 years inflation adjusted, or 25 years for the S&P 500 (Shiller data) or 26 years inflation adjusted. It looks like the 28 year figure was pulled out of a hat.
He says that "Starting in 1941 still encompasses a large part of those dark days in the market, and World War II". But by starting in 1941 (so that the initial high water mark is Jan 2, 1941), many of those "dark days", especially between 1943 and 1946 appear to be "happy" or "benign" days (new high water mark or within 5% of the most recent high).
Watch him turn dark nights into bright days.
Taking days at random strikes me as dubious. What's the chance that a day will be within 5% of the most recent high? Very good if the previous day was. Likewise, if yesterday the market was down 40%+ from its high, then the chances are much better that it will be down 40%+ tomorrow than if the market just hit a new high (it has never fallen 40% in a single day). While each day's movement may be random, one day's price is usually pretty close to the previous day's.
Certain things are obvious. Since the market has an upward bias, it will spend more time near highs than near lows. Just as obvious is that new highs will bunch - you're not going to hit a new high unless you're currently at or near a high. 2017 was a good example.
What are the odds of falling into a bear market if the market is already in a correction? Better than if it's hitting new highs. That's also obvious because it has a lot less to fall (a bear must begin as a correction). Conversely, if you're already in a bear market, what are the odds of entering a "second" bear market (i.e. falling 20% more)? Pretty small, because rarely does the market drop 40% or more.
So making use of any of this is tricky - too slow a trigger and you may smooth things out (miss the very bottom) but risk missing the rebound; too fast a trigger and you may get faked out and miss a rising market because it dipped for a week or a month.
Maybe not so thorough...thanks for your input.
Also, wouldn't a portion these new highs (referenced from the previous recent price) be coming off recent lows? If the market drops 20% in one day (Black Monday?) and on Tuesday the market rose to a "new recent high" and then over a number of incrementally higher highs (days...weeks...months...years) many more "new recent highs" would be necessary to retrace that 20% loss (with a 25% gain). In-other-words, markets may need more 'sunny days" to make up for the "dark days" because of the math - a 20% loss requires a 25% gain just to get back to even?
ISTM that losses often happen over fewer days and in larger negative increments...gains often happen over many more days and often in smaller positive increments.
"Recent high-water mark" could be the most recent local maximum, or it could be the last global high-water mark.
Think hills and valleys. A local high water mark would be the last hill you traversed.
But suppose all you cared about was "highest so far" (global maximum), and you traversed hills of 500', 1000' and 800'. The 500' hill would be your highest up to that point (a global high water mark). Then the 1000' hill would be a new high water mark (and the most recent one). When you got to the 800' hill, it wouldn't be a new high water mark. The 1000' hill would still be the most recent high water mark, with the 500' being an older high water mark.
FWIW, that's the way "high water mark" is used with hedge funds, and in Boston, where it was literal.
https://www.investopedia.com/terms/h/highwatermark.asp
http://whdh.com/weather-blog/historic-storm-to-historic-cold/
("major coastal flooding including in Boston. It was high enough to challenge the high level water mark of the Blizzard of ’78!")
Similar problem with his own term "happy day", that he defined as a day that "represented a new high-water mark." He noted that this is something "That works out to 5.4% of all days." Yet in his discussion he said that happy days occurred 36.3% of the time. He meant "benign" or "happy" days (0-5% below high water mark), but confused his own terminology.
He's provided food for thought. Though each of us may need to think it through from our own perspective.
CENSX, another oldy (widely touted in the 1980s and after (insurance-heavy)), took much much longer, till 1949.
There are a few others to look at starting summer 1929 but I did not delve.