Timepass's notes

[columbus]

Testing

--------------------BUY----------------SELL	
No. of contracts	Amt in Crores	No. of contracts	Amt in Crores
INDEX FUTURES	97181      2439.40	106947	2668.82
INDEX OPTIONS	452364    11243.89	438898	10932.87
STOCK FUTURES	153959    3865.47	167407	4184.91
STOCK OPTIONS	18295       455.24	17522	439.54

Better use HTML tag (<>) rather than (.) delimiter.If nothing is used the
entire text will be posted deleting all blanks in the line.

 

[timepass]

So, the FIIs are selling futures and buying options ?? PCR in index options stands at 1.16 today..

 

[timepass]

Well, well.. everyone was selling today in cash market. Let's see what the F&O numbers bring out.

FII trading activity on NSE and BSE in Capital Market Segment(In Rs. Crores)
Category 	Date 	Buy Value 	Sell Value 	Net Value
FII 	20-Oct-2011 	1468.86 	1940.86 	-472


Domestic Institutional Investors trading activity on NSE and BSE on Capital Market Segment

DII trading activity on NSE and BSE in Capital Market Segment(In Rs. Crores)

Category 	Date 	Buy Value 	Sell Value 	Net Value
DII 	20-Oct-2011 	719.91 		845.18 		-125.27

 

[timepass]

Bookmarking Post No. 534 in Uttamkamlesh's thread. Need to look closer at the type of M's and W's he is describing. Will try to co-relate these shapes with Bollingerworm, since Columbus too has described similar shapes.

 

[timepass]

Link to some of Uttamkamlesh's posts in "How to trade Commodities...." thread.
https://docs.google.com/open?id=0ByF-21VBnZ9bOTUwMmY2OGQtZjcxYy00YTI2LWE3ZmItZTgyMWFkZDFmYjhj

 

[trader.trends]

Timepass, Patterns like M and W stand out in the chart for the very reason that those were successful patterns, that is if you had traded them you would have made money. Failed Ms and Ws don't stand out at all. Try spotting failed patterns and you will see what I mean. So we tend to form a bias that Ms and Ws are profitable. When you want to back test this you will use only the successful patterns but ignore the failed patterns.

I know that we keep SL for failed patterns. My focus is that when we look at charts we don't see failed patterns as much as we see successful patterns. I hope you get the drift of what I am saying.

All such trading on pattern spotting suffer from "survivorship bias" . How will you deal with that??

 

[timepass]

I see what you mean. I have been looking at M and W patterns for quite some time now and have seen failed patterns. The ones described by Uttamkamlesh are typical because they include confluence of two different time frames. What I find interesting is that an M with higher right shoulder, and a W with a lower right shoulder may imply a strong continuation of the bearish/ bullish trend. This may be due to combination with a different time frame.

 

[adityasaraf007]

Timepass, Patterns like M and W stand out in the chart for the very reason that those were successful patterns, that is if you had traded them you would have made money. Failed Ms and Ws don't stand out at all. Try spotting failed patterns and you will see what I mean. So we tend to form a bias that Ms and Ws are profitable. When you want to back test this you will use only the successful patterns but ignore the failed patterns.

I know that we keep SL for failed patterns. My focus is that when we look at charts we don't see failed patterns as much as we see successful patterns. I hope you get the drift of what I am saying.

All such trading on pattern spotting suffer from "survivorship bias" . How will you deal with that??

I had observed the same "Survivorship Bias" while trying to Spot Wolfe Waves on Historical Data....

The only way to eliminate this bias, I guess, is by spotting such patters with the help of Bar Replay

 

[trader.trends]

I see what you mean. I have been looking at M and W patterns for quite some time now and have seen failed patterns. The ones described by Uttamkamlesh are typical because they include confluence of two different time frames. What I find interesting is that an M with higher right shoulder, and a W with a lower right shoulder may imply a strong continuation of the bearish/ bullish trend. This may be due to combination with a different time frame.

Don't trade them unless you are able to find atleast 200 such pattern and they give a positive return. Testing with smaller sample size and trading that in live ruins the account rapidly.

 

[timepass]

Of course, I am not yet trading the pattern shown by Uttamkamlesh, simply because I don't think that my software allows trendline from one timeframe to be shown in another timeframe. But it's an interesting concept.