Boolean algebra inside.
- Thread starter Kurgan
- Start date
Hi frenchfry,
I found that some of the stocks in your universe exceeded the float limit (i.e. between 5,000,000 and 60,000,000). Why did you include those stocks in your universe?
Thanks.
I am working on PVT too. I used the following criteria to get the universe:
My criteria:
1. 10 <= Px <= 50
2. 5,000,000 <= outstanding shares <= 60,000,000
3. average 65 days vol >= 200,000
4. RS percentile >= 75 where RS = (Close - Close.240)/Close * 100
My universe contains around 170 stocks.
I did not include the EPS in my universe because I found the EPS could vary largely with differenct data sources (e.g. finviz, yahoo, trade navigator). Even I didn't include the EPS, I think I could use the rank to filter the high quality stock.
Here is my hotlist generated on 1 Mar 2013:
DDD : Rank [4.23]
ENOC : Rank [5.21]
INFI : Rank [6.42]
NTE : Rank [4.84]
RPRX : Rank [4.60]
SGMO : Rank [4.30]
SLCA : Rank [5.77]
SRPT : Rank [4.88]
TVL : Rank [3.55]
UNXL : Rank [11.29]
If I understand correctly, we only take out the stocks which fulfill "minimum 20% gain in 5 cycles within 6 months". Then we check if the ranked stocks are in DU state and monitor those DU stocks. Once the vol, MACD and Stoch criteria are met, we enter long.
Please correct me if I am wrong. Thanks.
My criteria:
1. 10 <= Px <= 50
2. 5,000,000 <= outstanding shares <= 60,000,000
3. average 65 days vol >= 200,000
4. RS percentile >= 75 where RS = (Close - Close.240)/Close * 100
My universe contains around 170 stocks.
I did not include the EPS in my universe because I found the EPS could vary largely with differenct data sources (e.g. finviz, yahoo, trade navigator). Even I didn't include the EPS, I think I could use the rank to filter the high quality stock.
Here is my hotlist generated on 1 Mar 2013:
DDD : Rank [4.23]
ENOC : Rank [5.21]
INFI : Rank [6.42]
NTE : Rank [4.84]
RPRX : Rank [4.60]
SGMO : Rank [4.30]
SLCA : Rank [5.77]
SRPT : Rank [4.88]
TVL : Rank [3.55]
UNXL : Rank [11.29]
If I understand correctly, we only take out the stocks which fulfill "minimum 20% gain in 5 cycles within 6 months". Then we check if the ranked stocks are in DU state and monitor those DU stocks. Once the vol, MACD and Stoch criteria are met, we enter long.
Please correct me if I am wrong. Thanks.
My current problem is how to classify if the stock is in DU state. If I use 25% of average 65 day vol as the DU threshold, I then have very few trading opportunities. Because the stock rises again before reaching that DU threshold.
Therefore I read Spyder's equity journals and tried to figure out how to find the DU threshold.
At that time, Spyder introduced the upper and lower band DU levels. He did not use average 65 day vol to calculate DU. He used the lowest volume bar for each of the preceding 5 months and averaged those volume as the DU volume. I did the same thing. I found that spyder's DU method and 25% of average 65 day vol could have big difference.
E.g. BONT was in my hotlist on 1 Feb 2013.
DU (spyder's DU method) : 117,560
DU (25% of average 65 day vol) : 52,062
I am still working on the DU classification stuff.
Am I on the right track? or did I overcomplicate the stuff?
Therefore I read Spyder's equity journals and tried to figure out how to find the DU threshold.
At that time, Spyder introduced the upper and lower band DU levels. He did not use average 65 day vol to calculate DU. He used the lowest volume bar for each of the preceding 5 months and averaged those volume as the DU volume. I did the same thing. I found that spyder's DU method and 25% of average 65 day vol could have big difference.
E.g. BONT was in my hotlist on 1 Feb 2013.
DU (spyder's DU method) : 117,560
DU (25% of average 65 day vol) : 52,062
I am still working on the DU classification stuff.
Am I on the right track? or did I overcomplicate the stuff?
Quote from BenzMercedesSL:
Hi frenchfry,
I found that some of the stocks in your universe exceeded the float limit (i.e. between 5,000,000 and 60,000,000). Why did you include those stocks in your universe?
Thanks.
Thanks for catching that. This is the float criteria that I use in TN:
5000000 < Common Shares < 60000000
I also tried reversing the "<" into a ">" but it didn't make a difference. When I showed my setting to Jack he said he had the same result but he might have overlooked this.
As I don't know how this Trade Sense language works I think above criteria might be wrongly defined. In another language I would have defined it maybe like this:
(Common Shares > 5000000) AND (Common Shares < 60000000).
Which is the correct one?
I also saw that Spyder at one point changed how he calculates DU. He and others who discussed this topic in the past have more experience and therefore had their reasons to deviate.
However whatever I read from Jack it was always 25% of the 65day volume average. See the "One pager" for example.
But let's ask Jack.
Let me first ask you some questions:
How do you compute the Rank?
Where is your PVAD score in this process and what do you do with it?
In this thread Jack has not yet brought up the subject of MACD and Stoch. Also in the "One pager" there is no reference to it. Your process sounds familiar and I think Spyder's TN filter does everything and only spits out this one or few stocks per day which fulfill all criteria.
Like before I would prefer if Jack as the "final authority" of his method gives you his answers and point of view. Jack, please help.
Yo, Frenchy!
No,imagine dropping all this hershey crap and just start trading by using common sense.You`d have been amazed on how things actually went.