Intermarket analysis has many fans and many detractors. The fans believe in it because these relationships are logical and some of them have actually built trading systems using them. The detractors will point out that these relationships (breakdown) and most system don't work in the long run. I have been working with Intermarket analysis for well over a decade. I will try in this thread to share some of my knowledge with you on this topic. One issue that many of the detractors discuss is that these relationships don't work out of sample. One advantage I have is that I discuss many of them in public some over a decade ago. If they made money over the past ten years, guess what I think they worked. I hope I get a lot of questions on this topic. Intermarket analysis is one of my favor areas of technical analysis. In John Murphy's first book, published in 1991 on Intermarket analysis he used the crash of 1987 to lay out his Intermarket hypothesis. The problem is that until I built and published Intermarket based trading systems in 1994, no one had confirmed his work in a public forum. Many institutional traders used the concepts, but rules to mechanical trading systems, which used Intermarket analysis, were not generally publicly available. I developed a very simple concept for Intermarket based systems. For positively correlated markets If Intermarket is in up trend and traded market in down trend then buy If Intermarket is in down trend and traded market is in up trend then sell You can use various concepts to defined an up and down trend. In most of my work I used price relative to a moving average. For negatively correlated market we have as follows: If Intermarket is in up trend and traded market in up trend then sell If Intermarket is in down trend and traded market is in down trend then buy This is my introduction entry in this thread. I will include many examples as well as a historical view of the Intermarket research I have published over the years during the coming days and weeks.
Murray, I'm interested to see where you go with this thread. I often use intermarket relationships to form a view about the bigger picture, however, any models I've created to expoit these relationships have proved to be underwhelming at best. Invariably timing the entry proves to be the stumbling block.
I read John Murphy's book on intermarket analysis several years ago, and I became enamored with it at the time. Unfortunately, I was not able to use the concepts in a practical manner so as to either improve my entry timing or reduce the size of my initial protective stop. Quite possibly, I didn't try hard enough at the time.
Being a 20+ year "trading the markets veteran", my view.... (A) intermarket analysis is bogus, and (b) if you consider ANYTHING except *price* in the market you seek to trade, you will hurt your overall performance.
In this thread I will discuss my work in intermarket analysis over the years. I have written many intermarket articles Feb 1996, April 1998,March 2001,April 2001,July2001,Sept 2002 Jan 2001. Futures owns the copyrights so I can't give the articles away. I do have a deal with them for my book , here is the link for my book. http://www.tradersstudio.com/Defaul...alogID=5&psnavcmd=CatalogItemDetails&tabid=38 This contains over 100 of my articles from 1995 to 2004. Reprints from Futures are $5.00 each so , the book offers alot of material. If some of you have these article review them and get your questions ready.
Intermarket analysis can be though of as a form of fundamental analysis, without the lag. During this thread I will share my thoughts and research with you . I mention the articles because that give me a time stamp. If I wrote about it in 1998 and it still works then that means something.
I recall your name and remember reading your articles. Nevertheless, I still view intermarket analysis as "a fools errand", overall. I spent 20 years chewing on all the data and indicators.... and boiled it down to K.I.S.S.... You don't get paid for having your position "in tune with interest rates", or "volume", or "what Greenscam said", or "inflation", or "the ROC of the MACD Stochastic".... You only make money when your position is in tune with PRICE. Either it is, or it isn't. I recommend traders learn to trade off of "PRICE ONLY". They will make more money and do so with less risk.
One of the reasons that intermarket analysis gets a bad name is that many people who use intermarket analysis are simply trying markets shotgun and optimizing parameters. They're not thinking of why a given market should work and what logic reasons , based on the premise a given intermarket relationship would decouple. Intermarket relationships are not always constant, they go through periods of ebbing and flowing. What we want from intermarket analysis, is give us a statistical edge. We don't expected to be perfect, just give us an edge. Intermarket relationships, often times to decouple. That does not mean that the relationship is not valid, or are not reliable. In fact, if we do our research right we can understand why they're decoupling and even predict it. One classic example. Was there in 1993, both the CRB Index and the 30-year bond rallied that summer. These two markets are both to be negatively correlated. The problem was that interest rates, were falling, prices of bonds were rising so rapidly that this causes what I call an Intermarket inversion. I found with my research that 2-3 sigma moves often would cause these problems. The logic is that interest rates falling stimulated the economy and that caused the fear of inflation and commodity prices to rally. In my next entry in this thread I will discuss some Intermarket relationships I have been using for over ten years and wrote about 7 years ago which still work today. We will test them using my simple Intermarket Bond model, which uses a negatively correlated Intermarket. The code for it is below in TradersStudio basic '******************************** ' Classic Intermarket Bond system trading using a negatively correlated market ' TradersStudio(r) 2004 , All rights reserved Sub ClassicInterBonds(TrLen,InterLen) Dim InterAve Dim TrAve ' this system trade bonds using a negatively correlated market as independent1 InterAve=Average(Close of independent1,InterLen,0) TrAve=Average(Close,TrLen,0) If Close>TrAve and Close of independent1>InterAve then Sell("BuyEnt",1,0,Market,DAY) end if If Close<TrAve and Close of independent1<InterAve then Buy("BuyEnt",1,0,Market,DAY) end if end sub
Not my deduction, but rather a confirmation of what was stated in a newsletter long ago by some high-profile-at-the-time-guy.... I don't remember who it was... maybe Ian McAvity? I didn't want to accept it then, but never let it out of my mind. Today, I accept.