Market TIming Using Golden Cross

Research
21

Can look at some of these:
Industry = RTAPRL or Industry = RTDEPT or Industry = RTDRUG or Industry = RTFOOD or Industry = RTHOME or Industry = RTNONA or Industry = RTTECH

First use the basic industry, then start to look at sub-components.

Can also look at these sub-industries:
https://www.portfolio123.com/app/taxonomy/stock/usa/subindustry

Then, go through and find the stocks in the sub-industries that might make sense using the screener. So, areas like:
Footwear
Leisure Facilities
Shoes and Apparel
Motorcycle
Retail
Home Construction / Repair
Leisure Products
Auto Parts and Equipment
Home furnishing
Home building

Try the whole sub-index as a timer, then can also go through and pick the largest company with the longest history in each and use those tickers as timers - can also create a ‘blended’ consumer spending index with a custom list of stocks from 5-10 ‘representative’ industries and use that as a basket timer for middle class spending.

If you click on the sub-industry, can get a stock list - for example (From home furnishings):

Description Symbol Type #Stocks Cap (M) 1Day %* 1Wk %* 1Mo %* 3Mo %* 6Mo %* 1Yr %*
S&P 500
$SP500 Index 500 25,828,172 -0.55% -0.97% 0.26% 3.28% 8.61% 13.19%
Tempur Sealy International Inc
TPX Stock 1 2,775 -0.66% -3.65% -5.52% 7.60% 13.75% -21.60%
Dorel Industries Inc
DIIBF Stock 1 576 -1.54% -0.34% -6.39% -4.27% -20.01% -23.33%
La-Z-Boy Inc
LZB Stock 1 1,385 -3.10% -6.11% -6.99% -3.32% 0.13% 10.74%
Leggett & Platt Inc
LEG Stock 1 5,455 -1.25% -4.25% -7.22% -7.45% -4.12% -10.62%
Home Furnishings
FURNITURE SubInd 17 24,893 -1.58% -4.14% -9.56% -15.88% -18.68% -26.79%
Ethan Allen Interiors Inc
ETH Stock 1 527 -4.15% -4.34% -10.59% -20.60% -11.70% -34.63%
Mohawk Industries Inc.
MHK Stock 1 12,540 -1.76% -4.14% -11.87% -23.27% -29.62% -34.54%

So, something like MHK could be used. Or MKH plus LEG.

For homebulders, LEN and DHI, PHM and NVR

For flooring, ticker FND (floor and decor holdings)

Beauty Products - Avon and Revlon (ABP and REV)

Footwer- Nike

Fashion- Ralph LAuren (RL) and Michael Kors (KORS)

For motorcycles, something like HOG (Harley)

I don’t think using a single stock to time the market makes sense, though. But a basket of them might (across a few different categories).

22

Tom - I haven’t heard from you in a long time.

I’ve tried all of your suggestions without much success. But thanks anyways.

“I don’t think using a single stock to time the market makes sense, though. But a basket of them might (across a few different categories).”

There is no perfect indicator. Obviously, each individual stock will have its own issues that may be a contrary indicator to the overall market direction. I am just looking for certain stocks that tend to represent the market conditions and therefore can contribute to a pool of warning signs.

Steve

23

I agree with this. One needs to use mainly economic indicators, not stocks, for market timing.
Composite Market Timing Increases Returns And Reduces Drawdown (published 2 years ago)…
https://imarketsignals.com/2016/composite-market-timing-increases-returns-and-reduces-drawdown/

Comp Timer RSP-TLT.png
Comp Timer RSP-TLT.png586×585 65.1 KB

24

“I agree with this. One needs to use mainly economic indicators, not stocks, for market timing.”

Georg - The economic indicators that we currently have at our disposal are few and limited. They are highly optimized to get an improvement versus the stock market by market timing. The results are highly dubious due to extreme over-optimization.

** EDIT ** The equity curve that you posted is highly optimized and for the most part a backtest. I could do the same (if not better) by highly optimizing moving averages of certain selected stocks, but I have not chosen to do so. I am not promoting an agenda here, I am performing research.

Thus, more and independent indicators are necessary for making better stock market predictions. I view stock performance as one avenue towards achieving that goal. Certain stocks MAY provide evidence as to the state of the economy. In my estimation, there is as much value in pursuing this approach as such things as Google searches for “debt”. As I said in previous posts, stocks are fallible in this regards, but so are ALL other economic indicators. They contribute to the characterization of the economy and as such add value.

Take care
Steve

25

I don’t do much with market timing. So just a question.

Would using the moving average of a carefully selected “basket of stocks” differ from using the moving average of an ETF?

Don’t some of you use crossing-of-moving-averages of ETFs as a signal? I guess some do considering that is the original topic of this post. And Steve is right that the “Golden Cross” is pretty popular. Or is it the “Death Cross” we will be looking for?

Anyway, if you do use cross-overs of moving averages for ETFs then wouldn’t it be possible—in theory—to make a basket that is as good (or marginally better) than an ETF?

Just another thought: would a consumer discretionary ETF approximate what Tom is suggesting? Maybe even use relative performance of discretionary vs utilities and/or staples? It does seem to work in hindsight. Especially, if I am a little flexible when adjusting those moving averages, just a bit.

Of course, I could hear this type of discussion every day when I used to listen to CNBC and Bloomberg. So none of this is my idea. And maybe this was all just noise. I wouldn’t disagree with anyone who said not worrying about consumer discretionary spending is one of the good things about turning the noise off.

Maybe you just need to add bulk shipping into the mix?

Anyway, if you move in and out of certain ETFs or follow their moving averages then Tom definitely has a point—assuming a properly constructed basket of stocks. After all, even if the basket was just a large number of the stocks in one of the ETFs you believe is predictive……

For the record, I think Tom and the others who do this are often on to something. I just don’t do it myself (at this time). If I did, I definitely would not ignore Tom’s general idea. I might even buy one of his Designer Models based on an idea like this if he were given time to seriously construct his basket of stocks.

-Jim

26

It depends. How is the ETF weighted and rebalanced? If it uses the float weighted method (as do most which track S&P/DJ indices), then there would be differences. If it is equal-weighted, then how frequent are rebalancing/reconstitution periods?

I am of the opinion that using ETFs or attributed-based (e.g., GICS sub-sectors) custom indices as cyclical indicators are better options than individual securities. A basket of securities’ utility as an economic indicator would appear to have greater permanence than a single security’s. Moreover, there is less non-systemic risk (i.e., event-driven noise) when dealing with a diverse basket, and even less when it is size weighted.

27

I’m of the opinion that a few broad-based stocks that represent the economy well are better than an ETF. That’s my opinion.

Take care all

28

Steve, I tested the various combinations.

Investing in RSP when one OR the other Economic barometers 50MA > 200MA and TLT when BOTH are not yielded:

13.65% ann return
-31.12% DD
0.91 sharpe
13.39% SD
0.39 beta
10.55% alpha

The rule I used was:

Eval(SMA(50,0,GetSeries(“FDX”))>SMA(200,0,GetSeries(“FDX”)) or SMA(50,0,GetSeries(“MU”))>SMA(200,0,GetSeries(“MU”)),ticker(“RSP”),Ticker(“TLT”))

I assumed next day close, 0% slippage, and ran it on a daily basis.

Using the 2 Affluence barometers was not as successful, with a 10.85% annual return. Neither was using all 4, with an annual return of 11% and change.

I also tested the four combinations of either Economic barometer or Affluence barometer.

The only combination that bettered the two economic barometers above was MU and MAR:

Eval(SMA(50,0,GetSeries(“MAR”))>SMA(200,0,GetSeries(“MAR”)) or SMA(50,0,GetSeries(“MU”))>SMA(200,0,GetSeries(“MU”)),ticker(“RSP”),Ticker(“TLT”))

14.2% ann return
-34.11% dd
0.92 sharpe
13.58% SD
0.43 beta
10.87% alpha

Both the MU/FDX and MU/MAR combos went to TLT on Sept 25, FYI.

29

Miro - Sotheby’s is the lesser of the Affluence indicators. In any case, I have started simulating with going to cash instead of TLT, so I can see the timing better. I have several Hotel chains and resorts that perform well, and about 90% are currently bearish according to the golden cross.