Russell Death Cross: BS or valid indicator?

quote:

---

long-term investor

---

Same here. I honestly have no clue about the death cross or any other charting knowledge, but I did notice that since 08-09, the blue line and the red line have crossed twice, maybe a third time, and the markets have done pretty good since then.

I don't know what it is, but I do like the name.

I ran a few more simulations for different time periods and moving average terms....and yes, one can easily backtest to find an effective strategy by picking and choosing variables until it works

bottom line is that is seems to work in flat or declining markets, but fails in long term bull markets, which I'm sure everyone knew already, but it was interesting to see the actual results from different scenarios.

from 08-13 using 30 & 120 instead of 50 & 200 moving averages, it doubled return of buy and hold (70% to 30%)

I used to be a skeptic of technical analysis, and for the most part I still am and think that most of it is bullshite. However, I have been reading some stuff about Clifford Asness lately and his research into the momentum premium for equity, and it's interesting stuff.

I think most chartists are spouting mumbo-jumbo, but for a famous 50-200 rule like this, IDK anymore... maybe it's kind of valid. There are quantitative analyst who trade on momentum and make a lot of money.

Well. I tried to read the referenced paper. I really did, but included in it are the following formula and comment.

Rpi,t- r f ,t = api+ ßpi MKTt + vpi j=iwjVALj,t+mpi j=iwjMOMj,t+ epi,t

"In this paper we intentionally keep everything as simple as possible"

I was certain at that point I was somewhere I shouldn't be.

It seems the conclusion might be something like a hybrid momentum/value portfolio outperforms one of either. I didn't understand exactly how the combined portfolio would be constructed, whether it would include a combination of the best value and momentum stocks, or include the stocks that had both value/momentum characteristics. The latter sounds like it would have to be companies that were growing earnings very fast, but continued to be trading at high value, which sounds like a very good idea. I'm pretty sure the former (combo of value and momentum stocks) is what he was talking about though.

I'm not sure how this paper relates to the 'death cross' other than the death cross being a 'confirmation' that momentum is disappearing from a stock or index as there is no value aspect to it.

As I see it, the death cross is a cutesy name to an obvious observation; that price increases (earnings increases or expectations) have slowed to the point of stagnating. I don't know whether I think that is supposedly historically accurate due to human nature, business cycles, or the reality that systems (businesses, groups of businesses, or by economies) by their complex nature cannot continue to change (grow) at high rates and will develop to a state in which the aspects of the system which created the change (growth) have created a system which at some point falls into a consolidation stage in which it must pare off non-dynamic aspects before change can continue.

I'm pretty sure what I tried to say will only make sense to me

Anyway, since valuations of stocks seems to be more arbitrary than fundamental (based on wide variations of P/Es over time and the lack of correlation between P/Es to actual future growth), attempts to quantify human behavior seems as important as actual financial performance and competitive advantage in predicting a particular company's future price.

I managed a bank portfolio for a short time right out of college. I quickly found that the very same news one month that sent bond prices up would send them down the next. The human interpretation of data (mood, feel) had more effect on prices than the data itself. This being assumed, technical analysis can be helpful, though not for the reasons technicians assert; that the stocks 'must' do certain things if recent behavior fits some pattern. Technical analysis may work at times because the market, people, behave in a predictable way....sometimes.

soooo, does the 50/200 moving average cross mean something? Sure, it means that price increases have stopped for some arbitrary period of time. There's no possible reason they must or should fall in the near future. The infinite economic, political, and natural factors that drive prices are always changing. I don't think those variables are reducible to a simple moving average indicator. Nothing is that simple (ha ha).

quote:

---

I'm not sure how this paper relates to the 'death cross' other than the death cross being a 'confirmation' that momentum is disappearing from a stock or index as there is no value aspect to it.

---

Yeah, I am in the middle of slowly accumulating a thread on hedge fund capital introductions and alpha generating strategies, and so I naturally had this stuff on my mind and perhaps tried to fit it into tangential discussions.

"When all you have is a momentum hammer, everything looks like a momentum nail..."

But yeah, I referenced the paper just as an academic confirmation that the momentum effect is real. The Fama-French 3-factor model is well established in confirming that value stocks and small cap stocks outperform other classes and equity stocks, and Clifford Asness was famous for solidly confirming that playing momentum stocks correctly with quantitative strategies also outperforms, with rigorous and empirically repeatable results.

If that is true, then various technical analysis folk wisdom, such as the Death Cross, might also be true. That's all I'm really trying to say.

quote:

---

The Fama-French 3-factor model

---

By the way, although this is the standard model, Fung and Hsieh proposed a 7-factor model in 2004, but I don't think it really caught on.

It's another one of those things that I've been meaning to read about when I get more time.

I probably won't read that Asness article in a careful way until freaking November.

They are beyond me too at the moment.

I often get on kicks where I try to outline subject material that is just beyond my grasp, in hopes that it will help motivate me to learn it better. Sometimes it works, and sometimes.... well sometimes I just leave behind an unfinished mess.

I find that academic papers on subject matters like this can seem virtually unapproachable for months on end, but if you have some motivating factor that makes you come back to it and hack away at it again and again, bit by bit, then once your done with your little project, you come to "own" a lot of what the paper is all about, and become fluent in understanding the jargon, and also gain a common sense intuition for what it all means.

I think I'm still a few months away from all that, but hopefully I'll get there.