05-13-2014, 12:43 AM
(This post was last modified: 05-13-2014, 12:50 AM by earthtodan.)
Thanks for the comments guys! I know it takes some effort to digest this stuff.
Exactly. I have a Google Docs spreadsheet that calculates my allocations, streams prices and tells me in real time how many shares I need to add to (or subtract from) any position to bring my portfolio into balance. If I want to add a new position, I don't have to manually rebalance my allocations, the spreadsheet does it for me. This method takes some of the thinking off my plate and also frees me from round-number psychological targets like $1000's and %'s.
"Y" is simply the dividend in most cases. The dividend is given a 1.5x weighting in the formula. For example, GIS yields 3% as of today's close, so the yield would be given a weighting of 3 x 1.5 = 4.5.
One thing that bothers me a little is that "Y" varies with the stock price. If the stock goes down and the yield goes up, should that change the investment thesis? That could get a little messy. I'll have to sort that out. Ideally I'll come up with a target price separate from the conviction level.
"y" is the part of the equation I like the least and will probably change first. This is an adjustment factor for dividends that are likely to grow faster or slower than earnings. For example, QCOM is raising its dividend 20% per year on 15% EPS growth due to its massive cash pile and lack of debt, and management says they intend to stay on that trajectory for a while. This could also apply to tech giants like CSCO, IBM, MSFT, or any maturing company that is making a shift from growth toward value.
As a negative example, I think SDRL is likely to grow its dividend slower than earnings, so I need a way to adjust for that and not give a 1.5x multiple to the entire 11% dividend. In this case "y" could be a negative number on some part of the dividend to bring the weighting down.
However now that I think about it, the debt/equity ratio sort of captures this already. I could probably give more weight to that and get rid of "y".
I spot checked the results against my portfolio, and it came surprisingly close to the convictions I had already assigned. That is, after applying the "5" denominator, which has no meaning except to make the numbers fall into the 1-5 category. Anyway, I'm pretty happy that the formula approximates how I've come to think about the stocks I own. However, I'm not ready to hand the keys over just yet!
That's exactly what it is. Maybe I should call it "M" for macro view. It has pretty limited power to move the result.
I'll post an update when I make some sort of meaningful improvement.
(05-12-2014, 08:45 PM)Kerim Wrote: So if you had 30 stocks, with a total of 75 conviction points, then each point is worth 1.33 percent allocation. And a stock with a conviction score of 4 would get a 5.33 percent (roughly) allocation?
Exactly. I have a Google Docs spreadsheet that calculates my allocations, streams prices and tells me in real time how many shares I need to add to (or subtract from) any position to bring my portfolio into balance. If I want to add a new position, I don't have to manually rebalance my allocations, the spreadsheet does it for me. This method takes some of the thinking off my plate and also frees me from round-number psychological targets like $1000's and %'s.
Quote:And in the "C" formula itself, I have questions about Y and y. I think that the number of companies that have explicit payout ratio targets is relatively small -- so do you assign these yourself, and based on what criteria? And what do you mean exactly by "portion of the yield that is within the target payout ratio"? If the target is a payout ratio of 60 percent, and the actual payout ratio is 50 percent, then does Y get a value of 1.00?
Thanks!
"Y" is simply the dividend in most cases. The dividend is given a 1.5x weighting in the formula. For example, GIS yields 3% as of today's close, so the yield would be given a weighting of 3 x 1.5 = 4.5.
One thing that bothers me a little is that "Y" varies with the stock price. If the stock goes down and the yield goes up, should that change the investment thesis? That could get a little messy. I'll have to sort that out. Ideally I'll come up with a target price separate from the conviction level.
"y" is the part of the equation I like the least and will probably change first. This is an adjustment factor for dividends that are likely to grow faster or slower than earnings. For example, QCOM is raising its dividend 20% per year on 15% EPS growth due to its massive cash pile and lack of debt, and management says they intend to stay on that trajectory for a while. This could also apply to tech giants like CSCO, IBM, MSFT, or any maturing company that is making a shift from growth toward value.
As a negative example, I think SDRL is likely to grow its dividend slower than earnings, so I need a way to adjust for that and not give a 1.5x multiple to the entire 11% dividend. In this case "y" could be a negative number on some part of the dividend to bring the weighting down.
However now that I think about it, the debt/equity ratio sort of captures this already. I could probably give more weight to that and get rid of "y".
(05-12-2014, 09:28 PM)Dividend Watcher Wrote: Have you tried your formula on a test portfolio of 20-30 stocks? Were the results as expected?
I spot checked the results against my portfolio, and it came surprisingly close to the convictions I had already assigned. That is, after applying the "5" denominator, which has no meaning except to make the numbers fall into the 1-5 category. Anyway, I'm pretty happy that the formula approximates how I've come to think about the stocks I own. However, I'm not ready to hand the keys over just yet!
Quote:My one concern or question is about the 'c' (your conviction) variable. Is that to add your own 'gut feeling' about the stock or is that an out to push something higher up in the ratings because you 'like' the company.
That's exactly what it is. Maybe I should call it "M" for macro view. It has pretty limited power to move the result.
I'll post an update when I make some sort of meaningful improvement.