5/08/2007

Why 100% plus

On of the persistent question I get on email is why 100% plus, why not capture the first 100% plus itself. The simple answer is to reduce the trading universe and select the highest probability trade. More jazzy,stylish, and sounding like a black box, answer is, "Because I want to trade the long tail".

So here is how distribution of returns from 260 days looks like. Only for stocks with trading history of 260 days or more.

9 comments:

Unknown said...

I don't get your distribution. There is not negative returns for a 260-days sample?. The sample is the whole US stock market?.

x-> %return ?
y-> #stocks ?


best.

Pradeep Bonde said...

When you calculate moves from 260 days lows, how will you get negative returns?

Unknown said...

ok, I didn't understand the procedure.
thanks.

Unknown said...

Pradeep,
the time window you use to calculate the returns is the same for all the stocks?. Do you move the originial/final time slot of each stock in order to allow for all the stocks to have the same time to grow?.
I mean, if you starting point is a year ago, and final point is today I'd expect a bias because stock achieved 260-minimal little time ago have not had time to grow.
Best.

Pradeep Bonde said...

Stock can have min260 just a day ago and double next day. Look at the mortgage stocks like LEND and some others. The objective is to find momentum stocks, not do other analysis.

Unknown said...

Well if there is a stock that has gone down and never gone up... you should see -ve returns.

Are you sure that your algorithm is correct. There have to be some stocks that have gone down and only down.

One way to check this would be see when you record a low and the high in the previous 260 days. Then you make sure that the high date is after the low date

Pradeep Bonde said...

When growth is calculated from minimum close in 260 days the lowest value possible is 0. Which indicates close today= minimum close in 260 days. So values for growth from 260 days lows start from 0 onwards.

Unknown said...

ok.. so I understand it. But I have more questions

There are 139 stocks that showed 100% return. There are 89 stocks that showed 110% return.

That basically means that 89 of these 139 continue to give me additional returns and continue on the rise (approx 10% or more). Now my probability of hitting a winner (a stock that continues to go up) is 89/139.

Pradeep Bonde said...

What the distribution shows is a snapshot for a particular day, those 139 and 89 are two mutually exclusive sets.
So on that day 139 were up 100%
82 were up 110%
and so on.
So that probability calculation is not right.