A Guide To Contrarian Market Timing [View article]
Nice article, Robert. Like varan, I think this kind of careful, data-driven analysis is what we all want to find on SA. I look forward to reading more of your contributions!
Vanguard Stock ETFs With The Best Capture Ratios [View article]
Hi Bob, Another way to think about this issue is that, for an index ETF, the capture ratio compares the performance of that index to the benchmark. So the 5YR upside capture ratio of SPY is 99.37 and the downside is 99.56. All that tells you is that SPY tracks its index very well. Both ratios for any index fund relative to its index should be 100. So the question for the investor becomes, do I want to buy the S&P 500 index or another index that has performed better over the last five years?
If it were just a matter of measuring the volatility of another index relative to the S&P 500, then the capture ratios should show symmetry, that is, the upside and downside should be the same. If VIG has a 5YR standard deviation of 16 compared to SPY's 19, then we ought to expect that VIG's upside and downside ratios both ought to be 84. But they're not. The upside is 82.8 and the downside is 74.5. This means that investors were better off holding VIG rather than SPY over the last five years, because they won more on the upside than they lost on the downside.
The problem with using the S&P 500 as a benchmark is that it is not comprehensive enough. If Morningstar or other online sites used some kind of total market benchmark, then capture ratios would measure the upside and downside performance of a subset of the total market to total market itself, which would be a more valid statistic.
For instance, SPY tracks the S&P 500 and XLU tracks the subset of S&P utility stocks. Morningstar gives the capture ratios for XLU relative to the MSCI World NR USD index. It would be much better for them to provide the ratios compared to the index from which the stocks are drawn. Likewise, if every US stock index were compared to a US total market benchmark, then the capture ratios would be more useful. Paul is right that comparing a small-cap fund to a large-cap benchmark is somewhat misleading.
Now for the beta issue. All betas are measured relative to an index, so they have the same issues as capture ratios. Beta isn't better, it's just a different way of comparing two indexes. The problem is that beta doesn't tell you anything about asymmetry, which capture ratio does. The upside minus downside difference for capture ratios lets you compare the upside versus downside volatility of an EFT relative to some benchmark. To me, that gives the investor more information about which fund to buy. In this respect the downside capture ratio is more like the Sortino ratio. I want an ETF that not only has less volatility than the benchmark (lower standard deviation, lower beta), but one that has asymmetrical volatility: more upside, less downside. That's what the difference statistic measures.
That's enough for now. I gathered the data at hand to write another article on beta, which I didn't publish. I guess now I need to publish that article.
Thanks for your comment and the lively discussion, as always!
Vanguard Stock ETFs With The Best Capture Ratios [View article]
Hi Paul, On its Ratings & Risk page, Morningstar does provide capture ratios based on the most suitable benchmark for each particular ETF. The results appear just below the capture ratios based on the S&P 500. This allows for a more focused comparison with similar ETFs. Beta is likewise calculated relative to some benchmark, often the S&P 500, and thus suffers from similar methodological issues. On Vanguard's website, betas for their ETFs are calculated relative to the specific benchmarks they track, so they are all 1. But that isn't very helpful. For a retail investor who is thinking about buying something other than a broad market index ETF, capture ratios facilitate comparisons with ETFs based on other indexes. Capture ratios are indeed useful in measuring the performance of active managers against their benchmarks, but limiting their use to one application seems a bit restrictive. Thanks for your comment.
German 10-year bonds recently closed at 1.42% and TNX has been following German rates down since March 19. 1.5% is by no means out of the question for TNX.
How To Become A McDonald's Millionaire [View article]
Repeat after me: Past Performance is No Guarantee of Future Results. Thirty years is a long time. Things change. All extrapolations from historical data are guesstimates at best. I enjoyed the article but NEVER eat at McD's.
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
Thanks for the tip, ellena (and sorry for the bad pun). I still own some TIPs and may take profits soon. Like TLT, TIPs are now statistically overbought.
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
TLT closed at 121.17 today (5/14), with RSI at 72.79 and thus statistically overbought. TNX is at 17.88 and 10-yr rates at 1.76, getting close to the low end of the trading range of 1.72%. Chaos in Greece and the possibility of the Euro's collapse could drive TLT even higher and rates lower. German 10-yr rates closed at 1.46% today, and US rates have been quick to follow Germany's lead downward since 3/19. If US rates close below 1.72%, then 1.5% may be in sight. TLT may have upside to 124 or 125 in that case. But like you, Rich, I think the best place to watch all of this is from the sidelines until yields get above 4% again. Thanks for your comment.
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
I've moved into short duration corporate bond ETFs like VCSH, since I think intermediate term bonds may get hit hard if rates rise. On the other hand, the Fed may keep rates low until 2014 and flight-to-safety buying may propel Treasury ETF prices higher if things get out of hand in Europe. It's hard to read the tea leaves at the moment, so I've opted for safety rather than yield. Thanks for your comment!
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
Hi Rich, Yes, this may be a good time to lighten up on bonds and look at some solid dividend-paying stocks. Trouble is, some of those stocks are also overbought at the moment. However, some attractive ones do exist. Thanks for your comment.
Recent lows in TNX were 1.72% on 9/22/11 and 1.80% on 1/31/12. How far down is down? 10-year German Bunds were at 1.60% today and 10-year Japanese notes at 0.86%. We could reach the former, but the latter is probably out of reach without sustained deflation. Good luck!
I think we're near the bottom of the trading range here, which is down about 1.72% (I'm not at the right computer to access the precise number at the moment). Unless the economy slips into recession again, I think this may be a good moment to take some profits in appreciated bonds. I have a list of five solid utility stocks with good yields that I'm considering putting some bond money into when the current pullback has run its course. I also have some data ready for another article on this topic, and when I get some time I'll put it out there. Thanks for your comment and keeping up the conversation.
A Guide To Contrarian Market Timing [View article]
Vanguard Stock ETFs With The Best Capture Ratios [View article]
Another way to think about this issue is that, for an index ETF, the capture ratio compares the performance of that index to the benchmark. So the 5YR upside capture ratio of SPY is 99.37 and the downside is 99.56. All that tells you is that SPY tracks its index very well. Both ratios for any index fund relative to its index should be 100. So the question for the investor becomes, do I want to buy the S&P 500 index or another index that has performed better over the last five years?
If it were just a matter of measuring the volatility of another index relative to the S&P 500, then the capture ratios should show symmetry, that is, the upside and downside should be the same. If VIG has a 5YR standard deviation of 16 compared to SPY's 19, then we ought to expect that VIG's upside and downside ratios both ought to be 84. But they're not. The upside is 82.8 and the downside is 74.5. This means that investors were better off holding VIG rather than SPY over the last five years, because they won more on the upside than they lost on the downside.
The problem with using the S&P 500 as a benchmark is that it is not comprehensive enough. If Morningstar or other online sites used some kind of total market benchmark, then capture ratios would measure the upside and downside performance of a subset of the total market to total market itself, which would be a more valid statistic.
For instance, SPY tracks the S&P 500 and XLU tracks the subset of S&P utility stocks. Morningstar gives the capture ratios for XLU relative to the MSCI World NR USD index. It would be much better for them to provide the ratios compared to the index from which the stocks are drawn. Likewise, if every US stock index were compared to a US total market benchmark, then the capture ratios would be more useful. Paul is right that comparing a small-cap fund to a large-cap benchmark is somewhat misleading.
Now for the beta issue. All betas are measured relative to an index, so they have the same issues as capture ratios. Beta isn't better, it's just a different way of comparing two indexes. The problem is that beta doesn't tell you anything about asymmetry, which capture ratio does. The upside minus downside difference for capture ratios lets you compare the upside versus downside volatility of an EFT relative to some benchmark. To me, that gives the investor more information about which fund to buy. In this respect the downside capture ratio is more like the Sortino ratio. I want an ETF that not only has less volatility than the benchmark (lower standard deviation, lower beta), but one that has asymmetrical volatility: more upside, less downside. That's what the difference statistic measures.
That's enough for now. I gathered the data at hand to write another article on beta, which I didn't publish. I guess now I need to publish that article.
Thanks for your comment and the lively discussion, as always!
Vanguard Stock ETFs With The Best Capture Ratios [View article]
On its Ratings & Risk page, Morningstar does provide capture ratios based on the most suitable benchmark for each particular ETF. The results appear just below the capture ratios based on the S&P 500. This allows for a more focused comparison with similar ETFs. Beta is likewise calculated relative to some benchmark, often the S&P 500, and thus suffers from similar methodological issues. On Vanguard's website, betas for their ETFs are calculated relative to the specific benchmarks they track, so they are all 1. But that isn't very helpful. For a retail investor who is thinking about buying something other than a broad market index ETF, capture ratios facilitate comparisons with ETFs based on other indexes. Capture ratios are indeed useful in measuring the performance of active managers against their benchmarks, but limiting their use to one application seems a bit restrictive. Thanks for your comment.
CenturyLink: This Dividend King Has Huge Upside Potential Now [View article]
CenturyLink: This Dividend King Has Huge Upside Potential Now [View article]
The Peak Of Fear [View article]
How To Become A McDonald's Millionaire [View article]
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
Thanks for your comment.
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
Bunds, Bonds And The Outlook For Treasury ETFs [View article]
Yes, this may be a good time to lighten up on bonds and look at some solid dividend-paying stocks. Trouble is, some of those stocks are also overbought at the moment. However, some attractive ones do exist. Thanks for your comment.
BHP Billiton: Bond Substitute [View article]
Bond ETFs To Buy As Rates Rise [View article]
Good luck!
Bond ETFs To Buy As Rates Rise [View article]