Here is a daily chart showing the VIX in the upper panel and the SPX in the lower panel, with Bollinger Bands (click to enlarge):
The purple arrow pointing to Friday’s candle shows that the VIX almost gave a sell signal. The bulls weren’t quite carried-away enough to push it below the lower Bollinger Band, but it was very close.
In the last episode on January 11th (blue arrows), the VIX opened and closed beneath its lower BB. That was an extreme burst of bullish complacency. We don’t quite have that at the moment, and while the market could push higher from here, the odds do indeed favor the bears.
If you want to short the market here, look at the blue arrows again. It took the market over a week to roll over after a very strong sell-signal from the VIX. Back then, you had plenty of time to wait for an ideal entry point, though it is possible that traders are more skittish now.
Matt,
“It took the market over a week to roll over…”
I noticed that “lag” time too. I suppose it makes sense that the brains of the masses need to catch up to the move.
K,
Following up to your bulldozing tax story, I’m sure many have seen this one about the airplane Kamikaze:
http://tinyurl.com/yjz44pe
Stuff like this happens this time of year. But from what I’ve been reading in the news, there’s a grass-roots effort to protest Washington’s handling of spending and taxes.
Film at 11:00
As long as we get another ‘spike’ in the VIX to near 30 (or higher would be just fine too) in the near term, I’m happy. I can’t watch the market every day, every hour. Therefore, I prefer to swing trade and I gotta take my shots when I can get them by guessing at levels I want to buy at based upon TA. This blog helps me do that. Thanks.
George
I have not posted any backtest code in quite some time. I tired of testing it. However, I did some work this weekend based on your agreement about the FXI and SPY similarity. The 36MA you like so much appears to have decisive influence on things. I don’t use the same parameter settings as you for MACD or STO, but look at this code and run it on SD. If you need any help on backtesting let me know. I know it has been a while since we have bounced ideas around. I wont explain what I am trying to do here, just run the code and see what you think of the results. If I did my homework right, it beats a buy and hold strategy by a large margin. I only ran it on SSO, so try it on that, and we can see if it is portable.
Entry
(MovingAverage[MA,Close,2,0,60,1] > MovingAverage[MA,Close,36,0,60,2] AND
MovingAverage[MA,Close,2,0,60,2] < MovingAverage[MA,Close,36,0,60,2] AND
Stochastic[StocK,21,5,3,D] < 80) OR
(MACD[Diff,Close,8,13,5,D] > .02 AND
MACD[Diff,Close,8,13,5,D,1] < .02 AND
Stochastic[StocK,21,5,3,D] < 80)
Exit
(MACD[Diff,Close,8,13,5,D] < .02 AND
MovingAverage[MA,Close,2,0,60,1] < MovingAverage[MA,Close,9,0,60,1] AND
MovingAverage[MA,Close,9,0,60,2] < MovingAverage[MA,Close,2,0,60,2]) OR
(Stochastic[StocD,21,5,3,D] > 80)
This needs to be run in the 60M time frame and have detect every one minute selected.
String
Unofficial Problem Bank List increases to 617
617 is the areacode here !! I feel proud that unoficial problems have reached boston. LOL
Stringm,
I’ll try that code. I haven’t done anything new either. I’ll give it a shot.
Thanks
K,
If the problem bank list gets to your zip code, we’re in trouble.
I haven’t even ran strategydesk in a while.
George
Don’t bother with that code. I have erred somewhere because I can’t repeat today what I had yesterday. I must have tweaked something and did not save it. I will get back to you, sorry.
String
Dutch government collapses over Afghan troop mission, http://www.ft.com.
It is often said the first move after a FOMC annoucement should be faded. I guess that applied Thursday AH as well.
George
I figured out what I did. You probably don’t want to mess with running it anyway. I tested SSO back to 1/1/2009 and it returned 67% beating buy and hold of 42%. I was basically trying to catch the moves when price had been under the 36MA on the 60M, and was moving up and over. I had a different set of rules if the MACD on the daily was less than .02 then if it was over .02. I was exiting when the Daily slow STO was getting over 80. I plan on looking at something similar, but maybe with the turn on the 21PMA on the 60m. See how things act after the 36pma has increased for 2-3 bars, after it had been declining. That 36MA of your’s does seem to hold some significance for sure. I will post the corrected code if you do however want it.
String
Stringm,
That sounds like something I’d sure like to try.
I’ve figured out why these in-line decision-making “rules” don’t work, because for entries and exits, questions need to be asked FIRST before a decision is made. Like, how did I get here, now what do I need to do to exit.
Just too many variables of entries and exits to be near-perfect. I know that simple is usually better, but this decision-making isn’t as simple as that; getting in and out needs to be determined by the “conditions” that exist.
George
Here it is. Just run it on SSO and look at the trades. Run it in the 60m interval and unclick the overide. Click the intra bar test every one minute.
String
Entry
(MovingAverage[MA,Close,2,0,60,1] > MovingAverage[MA,Close,36,0,60,1] AND
MovingAverage[MA,Close,2,0,60,2] < MovingAverage[MA,Close,36,0,60,2] AND
Stochastic[StocK,21,5,3,D] < 80) OR
(MACD[Diff,Close,8,13,5,D] > .02 AND
MACD[Diff,Close,8,13,5,D,1] < .02 AND
Stochastic[StocK,21,5,3,D] < 80)
Exit
(MACD[Diff,Close,8,13,5,D] < .02 AND
MovingAverage[MA,Close,2,0,60,1] < MovingAverage[MA,Close,9,0,60,1] AND
MovingAverage[MA,Close,9,0,60,2] < MovingAverage[MA,Close,2,0,60,2]) OR
(Stochastic[StocD,21,5,3,D] > 80) OR
(Bar[Close,D] < Bar[Low,D,1]) OR
(MACD[Diff,Close,8,13,5,D,1] > .09 AND
MACD[Diff,Close,8,13,5,D] < .07)
Stringm,
Great. I’ll give it a shot.
Thanks