Tabell’s Market Letter – March 25, 1988

TABELL-S MARKET LETTER 600 ALEXANDER ROAD, eN 5209, PRINCETON, NEW JERSEY 08543-5209 MEMBER NEW YORK STOCK EXCHANGE, INC MEMBER NATIONAL ASSOCIATION OF SECURITIES DEALERS, INC (609) 987-2300 March 25, 1988 – – – …..A new- book ChaoS-;-lly-J ames…..Glf!tl!!lC is;-in'1'ii''''''''''w-. abstruse mathematics intelligible to the lay reader. It examines the recently emerged concept that the true shape of most objects in nature is far too complex to be represented by the familIar Euclidian geometry of hnes, planes. and cubes. Consider, for example, the problem of measuring a shoreline. It is easy to describe a seacoast's length by drawing a straIght line between two pamts on a map of that coast and measuring the mlleage between them. A typical coastline, however, contams coves and promontories. If we were to decide to include such features in our measurement, we would have. instead of a straight line. a zig-zag pattern whose total length would be somewhat longer. It would be possible. moreover, to arrive at longer and longer lengths simply by contmuously reducing the mterval between measurements. The process could be carried all the way down to the level of atoms. The pattern produced by this procedure was named, by its discoverer. Benoit Mandelbrot. a fractal. and It is interesting that one of the measurements involved in exploring such complex shapes was a price-tune series, which defmition. of course. includes the stock-prlce charts with which we are aU familiar. We have always considered that the task of the market techniCIan should be to bring order out of chaos. and recognition of the chaotic nature of such phenomena as equity prices underscores the difficulty of this task. One technique long used by market analysts in trying to quantify the phenomenon of stock-price changes involves the use of a filter. (Since we are recommending books, it should be noted that probably the most extensive work in thIS area has been done by Arthur A. Merrill and IS summarized in his book. Filtered Waves, published by Analysls Press. Wilton. Connectlcut). We were persuaded to make filters the subject of this letter by the fact that the market's latest high. scored a week ago. is a haIr above 20 higher than the October low. As our readers know, we have suggested in recent years that a 20 futer seemed to be a good rule-of-thumb for establIshing the eXIstence of a bull or bear market. Given Industrials It is a simple matter to apply a 20 fIlter to the Dow Jones to the 1929' One century. It faus. however. to cope with markets of We are a price-time series, and cycles should roughly resemble each other in both dimensions. During the 1930 – 1932 bear market. there were no fewer than four counter-trend upswings of 20 or greater. ranging from 22 to 56 days lfl length. Comparable swmgs occurred while the market was moving ahead between 1932 and 1937. It turns out that we can adequately describe the period from 1929 to 1942 by applying a 38 filter. This yields cycles, measured from low to low. of 789 trading days in 1930 – 1932, 1707 days in 1932 – 1938, and 1227 days from 1938 – 1942. These are, of course. magnitudes that one normally associates with major cycle moves. However. we are not done yet. A filter as broad as 38. when carried beyond 1942. recognizes only one 8071-day upswing from 1942 to January. 1973. Our rule-of-thumb. 20 filter handles this period better but still falls somewhat short. The 12 years from 1949 to 1961 never produced a 20 correction, and this era. when filtered at that level. appears as a continuous 3239-day rise. It becomes apparent that a more sensitive tool IS needed to measure the price patterns of the 1950's. By hmdsight. a 13 filter works pretty well, accounting with reasonable accuracy for what we believe to be the five completed cycles occurrmg during that time period—1942 – 1946. 1946 – 1949, 1949 – 1953, 1953 – 1957, and 1957 – 1962. These cycles ranged between 764 and 1304 trading days in length. For swmgs after 1962. our 20 filter appears to be appropriate, or. at least. had appeared so until recently. As we have repeatedly noted 10 thIS space, the 1273 days between August. 1982 and August. 1987 are an awfully long span for a single bull market. By utIlizing a smaller futer, we can recognize a 493-day cycle in 1982 – 1984. thus making the upswmg from the 1984 lows appear more normal. However, the serIOUS conceptual dIfficulty with this interpretation is that It suggests that a major bear market was completed in just 38 trading days between August and October of last year. lOB days have elapsed SInce that low. and the pessimist can claIm, despite the 20 advance, that what we are seeing is a bear-market rally. However. we are lookmg at a market that has behaved excellently. If one wishes to take an optimistic view. and not too badly, it must be admitted by even the most confIrmed bear. If the Dow remains. as we suggested m our year-end forecast that it might. in a range bounded by the 1987 high and low. we will have only prolonged a period which is dIffIcult to identify cyclically. A short-term position, therefore. IS the only Olle we can responsibly take. but. it must be emphaslzed. such a VIew must be an optimIstIc one. ANTHONY W. TABELL DELAFIELD, HARVEY, TABELL INC. AWT,ebh Dow Jones Industrials (1200) S & P 500 (12,00) Cumulative Index (3124188) 2009.45 263.39 3637.92 No statement or expression of opinion or any other matter herein contained IS, or IS to be deemed 10 be, directly or Indirectly an offer Of the soliCitation of an offer to buy or sell any security referred 10 or mentioned The matter IS presented merely for the convenience of the subSCriber While we believe the sources of our Informalion to be reliable, we m no way represent or guarantee the accuracy thereof nor 01 the statements made herem Any action 10 be taken by the subscnbershould be based On hiS own InvestlgaltOn and mlormaltOn Delafield, Harvey, Tabelllnc, as a corporallon and rts offIcers or employees may now have, Of may later take, poSlllons or trades In respect to any securities mentIoned In thIS or any future Issue, and such poSItIon may be dlfferem /rom any views now or hereafter expressed In thiS ar any alher Issue Delafreld, Harvey, Tabell Inc , whICh IS reglslered With the SEC as an mveSlmenl adVisor, may give adVice 10 Its Investment adVISOry and other customers Independently 01 any statements made m thiS or In ….ny O1.her Issue Further Informahon on any securr\y mentioned herem IS avadable on request