Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
here is the 2x result from wild_card's last logic:
Year % Return
1986 38.60%
1987 150.21%
1988 59.78%
1989 53.67%
1990 33.15%
1991 297.94%
1992 86.70%
1993 93.68%
1994 56.86%
1995 232.94%
1996 227.47%
1997 235.45%
1998 219.21%
1999 282.07%
2000 123.24%
2001 85.23%
2002 115.96%
2003 67.27%
2004 110.01%
2005 23.43%
86-95 110.35%
96-05 148.93%
20 yr 129.64%
I mean trinq makes its own higher lowand and ema5 simutaneously makes its own higher low. vice versa.
Use the attached chart and the example dates. "lower high" and "higher low" are defined as what is commonly used.
Jim
When trinq and ema5 make a hingher low, it's a sell signal.
When trinq and ema5 make a lower high, it's a buy.
sell on 12/13, buy on 1/1, sell on 1/12, buy on 2/6, sell on 2/16.
Jim
http://stockcharts.com/def/servlet/SharpChartv05.ServletDriver?chart=$trinq,uu[e,b]dallnnay[db][pc5!...
Yes, trinq had a sell signal last Thursday.
Jim
Thanks. I guess I need to get the TA_lib thing. I see a bunch of #NAME's.
Thanks again.
Jim
wild_card,
Thanks for the results. You are right, there is no consistant improvement.
If you send me the spreadsheet, I'll see if I can do some treaking over the weekend. (djchen@iastate.edu)
Jim
wild_card,
I pulled up the charts of compq for 87, 90, 98, 2000, and 2001. By eyeballing, it seems that an override of long signals using {ppo(5,18)<-3.2 and falling} should help significantly.
I did not try the bull side, but suspect another ppo setting should help also.
Any thoughts?
Jim
Thanks. I did not see this before I posted the previous one.
Jim
I guess I was not clear.
I was not suggesting a new setting for the ppo. As I suggested earlier to tclark1, I think using a "super bull" flag to suspend any short action during bull mode, and a "super bear" flag to suspend any long action during bear mode can help reduce drawdowns.
You can define "super bull" by {ppo>3.2 and ppo rising}, and define "super bear" by {ppo<-3.2 and ppo falling}. The number 3.2 and -3.2 can be treaked to other numbers.
Then in your bull mode logic, whenever you get a short signal, check to see if "super bull" is flaged. If not, go with the signal; if yes, don't short (go to cash instead).
In the bear mode logic, whenever you get a long signal, check to see if "superbear" is flaged. If not, go with the long; if yes, don't long (go to cash instead).
Hope the above is clear and look forward to seeing the results.
Thanks.
Jim Applesmith
wild_card,
Here is another suggestion:
If ppo >3.2 and rizing, disable any short in bull mode
If ppo < -3.2 and falling, disable any long in bear mode.
The number 3.2 can be treaked. The positive side and negative side do not have to be the same.
Jim Applesmith
Did you include the super bull/super bear disablement for the RSI overrides?
Jim
This reduces the drawdown: during super bull/bear mode, diable override.
Super bull: ppo(8,18,5) > 2 and rising
super bear: ppo < -2 and falling.
The number "2" can be treaked.
Rising = {today's value > yesterday's value}. The other way is falling.
Jim
Here is a different suggestion:
If ppo(8,18,5)'s signal line is {(>-1) and rising} or {>1}, use bull mode rules;
If ppo(8,18,5)'s signal line is {(<1) and falling} or {<-1}, use bear mode rules.
Jim
ujpix's price is determined not by the actual nikk price of the day, but by the perceived next morning's close projected by NYSE traders at 4:00pm NY time.
For example, even though yseterday's NIKK gain is not large, the NYSE traders believe today's NIKK gain will be large, which leads to a large gain in yesterday's ujpix price. Today's ujpix price will be dependent on what NYSE traders think tomorrow's NIKK gain will be. Therefore, even though NIKK had a large gain today, it does not mean ujpix will have a gain, since today's nikk gain was already factor into yesterday's ujpix price.
The above is based on the prospectus and answers from profunds.
Hope it's not redundant info, and best luck to you.
Jim
Thanks for sharing your thoughts on ujpix. Yesterday should be a good buying opportunity for you according to your plan. Good luck to you and me.
Jim
Thanks.
Jim
If one plots ujpix in the background of NIKK over one year, the two plots track nearly perfectly. Yet, from day to day, the mismatch can be large. For example, on 1/12, the NIKK gained 0.5%, but ujpix dropped 1.3%. Can such discranpacy be used for benefits? Does any body know how to predict the daily close of ujpix based on intraday data?
Thank.
Jim
MFB and the board,
The daily NAMO at stockcharts is only available after market close. Is there any web site that gives the estimated daily namo using intraday data?
Thanks.
Jim
Even though I don't trade, I truly appreciate your hardwork and your willingness to share and to teach.
Jim
I just want to say that I greatly appreciate your sharing such a chart before the market close.
jim
Nocona,
Price line crosses SMA2 if and only if price becomes lower than previous hourly close in up trend or higher than previous hourly close in a down trend.
So there is no need for the SMA2, nor do you need to switch to line graph. Simply compare the current price against the previous hourly close and you get identical info.
Great system. I wish I had the time to follow your system.
Jim
That's interesting infomation. Is the daily relative strength score available over a past period of time? By plotting each fund's historical RS, one can identify when a strong performer is beginning to lose steam or a laggard that has just bottomed out.
Just want to say that I appreciate your posting the SI plot, especially the intraday update.
jim
Metal,
I really appreciate your posting the daily update before the close. Does stockcharts provide intraday SI and OSC?
Thanks,
jim
Your NAA50 plot will be more revealing if plotted on a linear scale (no log, no percentage).
http://class.ece.iastate.edu/djchen/Picture1.emf
1. your board and your charts are great.
2. you forgot to sort the data for your distribution charts.
dj
Great work. Do you have a chart showing historical infor in both a rising and falling market? Thanks.