Miles P. Jennings, Jr.'s  Instablog

Determining portfolio weights for a set of large-cap companies (or for an ETF, Fund, or Index) can present analytical challenges.  However, before delving into company-specific results and their respective projected earnings, a simple mirror-weighting of the portfolio can serve as a valuable starting point in the stock allocation process. 

For example, let's assume a portfolio of $100,000 with the following 5 companies to be held: AAPL, GOOG, IBM, MSFT, ORCL.  The sole weighting factor of Earnings is specified.

The MirrorWeight Widget is used to calculate the portfolio weights.  To create a mirror-weighted portfolio yourself at www.mirrorweight.com, you must enter only the portfolio's size, the stock symbols, and your choice of weighting factor(s).  

Here are the calculated portfolio weights---mirroring the companies' earnings:

Mirror-Weighted Portfolio

Stock	Shares	Price	Position  Value	Portfolio Weight	P/E	P/B
GOOG	21	548.29	11,629	11.63%	35.43	5.18
AAPL	67	197.37	13,129	13.13%	31.38	6.43
IBM	267	120.65	32,163	32.16%	12.81	10.54
MSFT	773	28.59	22,091	22.09%	18.65	6.23
ORCL	960	21.87	20,988	20.99%	19.63	4.24

Note: Showing just whole shares and rounding causes slight variance from true data.

To check the math and demonstrate the MirrorWeight method, we only have to invert the P/E ratios and calculate those % as follows:

MirrorWeight Check

Stock	P/E	E/P	E/P to Total E/P
GOOG	35.43	.0282	11.6%
AAPL	31.38	.0319	13.1%
IBM	12.81	.0781	32.2%
MSFT	18.65	.0536	22.1%
ORCL	19.63	.0509	21.0%
Total		.2427	100%

The above example uses earnings as the factor, but you may also specify book value or a combination of earnings and book value.  In looking at the above example, IBM has the highest portfolio weight since its P/E Ratio is lowest.  And note that the portfolio weights in the first table are exactly the same as the "E/P as % Total E/P" in the second table.

Other MirrorWeight Widgets and their programming code may be found at www.mirrorweight.com.  

Here are two discussions about the MirrorWeight approach:
Method:  www.mirrorweight.com/ratios.htm
Market Example:  www.ownershipview.com/method.htm

MirrorWeight Widget financial data is provided daily by EDGAR Online, Inc. and includes about 10,000 companies.  Data is in as-reported XBRL format. The portfolio display is via OV Engine. 

I own OV Metrics, LLC (private) which company operates MirrorWeight as an operating division.  I am long EDGR.