Radiomuse
288 posts
msg #91133
- Ignore Radiomuse |
4/13/2010 10:34:17 AM
RDN and MTG are the only 2 I ended up getting into - RDN is about flat and MTG up about 1%.
|
papagatorga
124 posts
msg #91184
- Ignore papagatorga |
4/14/2010 5:08:36 PM
Kevin, I wasn't complaining. Did exactly what you said, run the filter and chose my stocks. Just wondered if this forum had been shut down. Thanks for your work and effort. PS, I've invested twice, lost both times. First time set a stop loss, it was triggered, then the next day I would have gotten my 5%. Second trade hit my stop loss also. I have had too many large losses so I am afraid to trade w/o a stop loss. I know I know FEAR and GREED. FEAR is also troubling me with Market at Open orders.
|
Kevin_in_GA
4,599 posts
msg #91198
- Ignore Kevin_in_GA |
4/14/2010 9:35:19 PM
No problem - often my slightly sardonic comments fail to fully translate into posts.
This approach, based solely on past probabilities of hitting a target, does not guarantee anything. In some ways as the frequency of 5% hits keep going, the likelihood of a miss must increase (the hitting streak always ends sometime).
The statistics I posted in the first post were based on only trading the top ranked stock - no further analysis used.
|
chetron
2,817 posts
msg #91203
- Ignore chetron |
4/15/2010 6:55:04 AM
kev, if statistics have any validity, with each passing hit should be increasing the likelihood of it happening again, not that it will fail.
being due is a fallacy.
hth
|
Kevin_in_GA
4,599 posts
msg #91205
- Ignore Kevin_in_GA |
4/15/2010 8:31:01 AM
My thinking on this is based on observation of "streaks" - the baseball player who has hit at every at bat for the last twenty times is pushing the historical statistics of continuing said streak.
In some ways it depends on which way you look at it - is each week a coin flip where the likelihood of continued heads decreases over time, or it it more like when Michael Jordan gets to the line for a foul shot (where his historical likelihood of making the shot is a pretty solid indicator of how he'll do this time)?
|
chetron
2,817 posts
msg #91206
- Ignore chetron
modified |
4/15/2010 8:56:04 AM
or maybe he is just getting better???????
with steriods maybe = )
|
LebowskiUrbanAchievers
56 posts
msg #91207
- Ignore LebowskiUrbanAchievers |
4/15/2010 9:26:49 AM
Other than making a mental note or manual chart on how many times a stock has been hit on the filter. Is there a way to program display how many weeks a stock has hit the filter?
|
Kevin_in_GA
4,599 posts
msg #91208
- Ignore Kevin_in_GA |
4/15/2010 9:31:01 AM
The truth is that what we are looking at is VOLATILITY, rather than PERFORMANCE. In that regard, the likelihood of high volatility can continue unabated for long time frames. There is also an arbitrary weighting factor I have included here to more heavily weight recent volatility over older data, which will penalize stocks that missed the target once in in the last month.
Clearly the goal here was to select stocks in a potential streak and ride it.
|
Kevin_in_GA
4,599 posts
msg #91209
- Ignore Kevin_in_GA |
4/15/2010 9:36:46 AM
Hit the filter or hit the target?
The target hit rate is drawn on the bottom of the chart, along with how often it also hit the 10% stop loss. As for how many times in the past year it has been a part of the filter output ....
you could add the following lines
set{freq, count(reward > 2.79, 52)}
add column freq
This should let you know how many times in the last year it had a reward score of 2.8 or greater, which would include it in the filter output for that week.
|
Kevin_in_GA
4,599 posts
msg #91234
- Ignore Kevin_in_GA |
4/15/2010 8:05:44 PM
By the way, the list for this week was CPE, CIGX, CDIV, PCBC, PMI, CNLG, and BGP.
Right now, all but CIGX and CNLG have hit their 5% target.
|