If you bought AGNC after the spo, your basis is probably around 29.50. AGNC closed at 30.57 yesterday or a 1.07 gain. The dividend is 1.25 so I think some are locking in the gains since we still have a long run to the next ex date. Compound the Spain and spreads I think a lot of this is just profit taking.
My feelings exactly. Just sold 5000 shares at 30.66 and hope to buy back soon if any drop on Monday after earnings. Maybe just profit taking like myself after the run up recently? Obama is trying to do an end run around congress to reduce mortgage principal which could affect the MBS market? Just a guess, I am not an expert on these matters.
I think that AGNC is not closely correlated with the market. Some stocks tend to follow the general market up and down like a puppy dog, but the MREIT's have more to do with interest rate activity. If anything, some people will take money out of AGNC today, and invest in other stocks with the market up today. Just look at AGNC on it's own merits; don't worry about the market so much for that analysis.
It is not that hard to 'run' a correlation study.
If u hv a spreadsheet program that accepts (will input 'CSV" values [comma seperated values] it is very easy.
Within the Microsoft enviornment (spelling?)__u download from Yahoo historical data in "CSV".
Now open Excell (or equivalent). Create a spreadsheet.
You just hv 2 choose which parameter is better 2 b horizontal versus vertical.
Copy and paste your downloads in2 columns and rows.
Now apply the 'Correl' function__to calculate the correlation between your column & row__or row & column.
As a side note__we hear all the time "talking heads" verbousing (spelling?) about a correlation between different asset classes. I encourange all to 'run' their own correlation analysis. You may be surprised by the results.
""""Gold & Oil go down if the US$$ dollar increase?""""
""""The Renimbi goes up if the US$$ decrease?""""
Traesuries increase when 'Junk Bonds" decrease?""""
It is eaier than ever B4 to run an analysis.
It is not that difficult.