simba wrote:If you build on this spreadsheet, you may wish to double check the Vanguard returns and especially the Non-Vanguard returns.
And I flipped a couple of 2008 to 2009 switches, but not sure if I got them all.
Paul Douglas Boyer wrote:I have already found an error: I used Average instead of CAGR in the charts.
Here is a link to a fixed version. I have also gone through Vanguard's site to add in their 2009 returns. If you build on this spreadsheet, you may wish to double check the Vanguard returns and especially the Non-Vanguard returns.
And I flipped a couple of 2008 to 2009 switches, but not sure if I got them all.
I put a chart in the Lazy_Portfolios_85 tab, feel free to delete it since it does not include all of the portfolios in the table.
Paul - I did not use your spreadsheet as such. I had already made a few changes to the latest (local - unreleased) version rev8j with the 2009 returns. I did notice that you had an error in your version, but in my version I was using the CAGR. I meant to post it later tonight but I guess you caught the error.
Thanks for the updated spreadsheet. It is clear a lot of work went into this. I took a look at the OpenOffice version and it seems to be missing the 1985-2009 data for a few mutual funds: BRSIX (J80), PCRIX (R80), and PIGLX (AM80).
I assume the goal is to track total returns for the calendar year, which would include dividends/splits, etc., as Morningstar does. If so, one 2008 number might be wrong as well:
BRSIX (see the performance tab for the fund at Morningstar for this data):
2008: -39.5 (not -41.74?)
2009: 26.0 (missing)
PCRIX:
2008: -43.3 (correct)
2009: 39.9 (missing)
PIGLX:
2008: -2.7 (corrrect)
2009: 17.2 (missing)
The 1972-2009 data should be updated similarly, I assume. Minor aside: Lazy Portfolios cell C1 references 2008.
Just adding a nod here that I've been playing with this spreadsheet during its many incarnations and it continues to improve. My portfolio is rather uncomplicated but it allows the freedom of many wonderful 'what-if' scenarios.
Now, if we could only download the 'crystal ball forward-gazing' spreadsheet!
I noted that the return you show for VBMFX for 2009 is 6.79% in your two data sheets. However, the correct return I believe is 5.93%. VBIIX return for 2009 was 6.79%. Is it your intention to change the ITB category to use VBIIX?
Unfortunately, I only have a bond index fund available (DBMIX) that tracts the Lehman Brother's Aggregate Bond Index....what would be the appropriate equivalent?
I apologize if this has already been answered in the 7 pages of replies - is there any plan to get international small value data into the spreadsheet?
I understand that prior to VINEX this class was somewhat un-investable and the various ETFs have but a few years of data. Are there any index return data for this class out there? Perhaps one could subtract 0.5% ER from index returns and use that as a proxy for the class.
Thanks again. This spreadsheet is a wonderful resource. Chuck
Thanks. Glad you like it.
RTR2006 wrote:Just adding a nod here that I've been playing with this spreadsheet during its many incarnations and it continues to improve. My portfolio is rather uncomplicated but it allows the freedom of many wonderful 'what-if' scenarios.
Now, if we could only download the 'crystal ball forward-gazing' spreadsheet!
Thanks for a wonderful, flexible tool.
RTR
RTR - Thank you for the kind words. Good luck in finding the 'crystal ball forward-gazing spreadsheet'
jamacq wrote:I noted that the return you show for VBMFX for 2009 is 6.79% in your two data sheets. However, the correct return I believe is 5.93%. VBIIX return for 2009 was 6.79%. Is it your intention to change the ITB category to use VBIIX?
Jeff - Thanks for catching it. The error was an oversight and unintentional. I'll correct it in the next revision.
kb0fhp wrote:This is excellent - and very easy to use.
Unfortunately, I only have a bond index fund available (DBMIX) that tracts the Lehman Brother's Aggregate Bond Index....what would be the appropriate equivalent?
Both DBMIX(Dreyfus Bond Market Idx) and VBMFX (VG Total Bond Mkt Idx) track the Barclays US Aggregate Bond Index (formerly Lehman Brothers Aggregate Bond Index) so you can use the VBMFX. Keep in mind the ER for DBMIX is .4 where as that for VBMFX is .2
Anagoge wrote:Thanks for the updated spreadsheet. It is clear a lot of work went into this. I took a look at the OpenOffice version and it seems to be missing the 1985-2009 data for a few mutual funds: BRSIX (J80), PCRIX (R80), and PIGLX (AM80).
I assume the goal is to track total returns for the calendar year, which would include dividends/splits, etc., as Morningstar does. If so, one 2008 number might be wrong as well:
BRSIX (see the performance tab for the fund at Morningstar for this data):
2008: -39.5 (not -41.74?)
2009: 26.0 (missing)
PCRIX:
2008: -43.3 (correct)
2009: 39.9 (missing)
PIGLX:
2008: -2.7 (corrrect)
2009: 17.2 (missing)
The 1972-2009 data should be updated similarly, I assume. Minor aside: Lazy Portfolios cell C1 references 2008.
I thought I added the PCRIX and PGLIX but looks like I inadvertently missed it in the final revision. I have added them to the next revision.
Where did you get the 2009 BRSIX returns? There was a mismatch for the 2008 returns for BRSIX between Bridgeway and Morningstar websites. I used the returns listed on the Bridgeway website.
simba wrote:Where did you get the 2009 BRSIX returns? There was a mismatch for the 2008 returns for BRSIX between Bridgeway and Morningstar websites. I used the returns listed on the Bridgeway website.
I was referencing the Morningstar data. The latest Bridgeway audited prospectus (click "Get a Prospectus" in your link above) also agrees with Morningstar (page 21/23).
It isn't a big deal either way, but I suspect the Bridgeway web site just has an error.
Hi,
Each time a new backtest spreadsheet comes out I add a row to show the returns for the worst year. Closest that I can get to a maximum drawdown given annual data. I find this to be a very important and useful piece of data when evaluating a portfolio. Perhaps it could be added to the next version?
Thanks to all who contributed to this spreadsheet, It is invaluable.
FYI for any Mac users: I downloaded the Excel version of the spreadsheet. I don't have Excel on my computer, but was able to open and use the spreadsheet in "NeoOffice", an Open Office based suite. The calculators work, but the graphs don't seem to. I have NeoOffice 2.2.4
My Mac is an Intel-based Mac Mini running OS 10.5.8
Interestingly, NeoOffice could not open the Open Office version of the spreadsheet. There is a newer version available than the one I have installed. Maybe the spreadsheet would work with that.
I am having difficulties adding a new fund to the Backtest-Portfolio-returns-rev9b.xls spreadsheet.
The README tab says...
To add returns for additional funds, use the appropriate worksheet (Data_85_09 to add returns for the years 1985-2009. You only need to modify the cells with -
This will prepopulate the Fund Name and ticket symbol in the rest of the worksheets and then add the yearly returns.
I am adding VFICX (Vanguard Interm-Term Investment-Grade) to the Data_85_09 tab column AQ.
I added this info...
AQ1 = Interm-Term Invest-Grade
AQ2 = VFICX
AQ3 = 0.24
AQ4 = 1993
I got excited, then I questioned my recency bias, when I realized my own simple buy-and-hold, slice-and-dice play off of the UB&H with some modifications ... beat every single portfolio in the spreedsheet :S
Hi,
It's suprisingly easy to curve fit to the past, but backtesting is far from useless. I use backtesting for active trading and the way I look at is I am already a step ahead of the many many people that either have no strategy or use a strategy that never worked. At least I know the strategies I use worked at some point in time. Next I make sure that my strategy or allocation makes basic sense and is not relying on some unusual performance (eg. gold in the 70's, t-bonds in the 80's, emerging markets in the 70's thru 90's at a time when access to this class was generally not available or would have been much more expensive). The correlation values in the spreadsheet are useful. While they will break down in a bear market it is still useful to invest in uncorrelated assets.
DP wrote:Hi,
It's suprisingly easy to curve fit to the past, but backtesting is far from useless. I use backtesting for active trading and the way I look at is I am already a step ahead of the many many people that either have no strategy or use a strategy that never worked. At least I know the strategies I use worked at some point in time. Next I make sure that my strategy or allocation makes basic sense and is not relying on some unusual performance (eg. gold in the 70's, t-bonds in the 80's, emerging markets in the 70's thru 90's at a time when access to this class was generally not available or would have been much more expensive). The correlation values in the spreadsheet are useful. While they will break down in a bear market it is still useful to invest in uncorrelated assets.
Annual returns are a bit coarse. To see peak to trough issues you really need something like monthly data or a chart. Don't fall in love with the spreadsheet.
Thanks so much for making this spreadsheet available. I just found it and have been slow to understand the importance of international and emerging markets, so this helps a lot with seeing their effect.
I have been playing with changing the weights of combinations of TSM/VISVX/VGSIX/VEIEX/VTRIX and I think my "ideal" is 15%/30%/15%/15%/25%. Just one observation which surprised me, I think the Total un-rebalanced nominal almost always exceeds the rebalanced value ie While I see plenty portfolio reduction in Std Dev. I don't think I see any balancing benefit to return despite some individual correlations < 0.5... I also tried up to 5% VUSTX.
I was hoping that someone could explain my mistake in expecting a balancing boost.
SDBoggled wrote:Thanks so much for making this spreadsheet available. I just found it and have been slow to understand the importance of international and emerging markets, so this helps a lot with seeing their effect.
Thanks
Here's to helping speed things up for you: try adding in some gold also.
You will find that it increased the returns and decreased the risk.
Where are the bonds? Cash? I like the Harry Browne Permanent Portfolio for money you can't afford to lose.
SDBoggled wrote:
I was hoping that someone could explain my mistake in expecting a balancing boost.
Thanks
It depends on the market. 2/3 of the time it is going up, and 1/3 time it is going down. If the market is going up, and stocks have increased ,then cutting back on stocks can hurt future returns, if the market continues its raise.
As a result, not reblancing means the returns will increase. That is because your AA will be higher for equities, over time.
However, you are increasing your risk. While rebalancing does enforce a sell high buy low strategy. As you point out SD is lowered. I would expect that the Sharpe ratio should be higher, due to the buy low,sell high, while controling risk.
Thanks for taking the time, your explanation makes a lot of sense and does seem obvious once you spelled it out. I am shocked and disappointed how often I have trouble with such basics... makes me doubt that I "really know what I am doing".
So I revisited the model and saw that Emerging markets had gone from 15% to just over 50%... I had totally overlooked the "ending allocation if not rebalanced column".
Just drag-copy the AS99 cell down to the AS100 cell.
This will fix the bottom left graph on the Portfolio tab. If you'll compare the bottom-right and the bottom-left graphs you will see that the bottom-left graph stops at 2008 instead of 2009.
Mike
As Merton Miller, a Nobel laureate at the University of Chicago, puts it, "I'll never understand why they call bonds 'fixed' income."
Thanks for taking the time, your explanation makes a lot of sense and does seem obvious once you spelled it out. I am shocked and disappointed how often I have trouble with such basics... makes me doubt that I "really know what I am doing".
So I revisited the model and saw that Emerging markets had gone from 15% to just over 50%... I had totally overlooked the "ending allocation if not rebalanced column".
Thanks again.
It's a common misunderstanding, I think, in part because of the language used. We think of 'bonuses' as being financial in nature - i.e. a 'bonus' is something extra we get paid. In this case, it has a very different meaning - the 'bonus' is reduced volatility, which in general reduces yields. The key of course is that most people want to reduce, not increase, risk over time, hence rebalancing makes sense for most portfolios.
"In the absence of clarity, diversification is the only logical strategy" -= Larry Swedroe
Hi,
I was looking at the correlations on the data_85_09 tab and noticed some discrepancies. I wanted to see the name of the fund so I added in the list of fund names and symbols from the Portfolio tab in front of the existing fund symbol. In the list I see some duplicates and some mismatches, and I am not clear how the data is pulled in so I'm not sure which is correct.
See for example VWNDX, VWNFX listed multiple times, mismatches in the health care fund and the energy fund seems to be missing.
This is a remarkable tool, and I hope Simba comes back to update it, but even as is it is very valuable for backtesting portfolio alternatives.
What I've learned from this is that a few funds, like in Taylor's 3 or 4 fund portfolios, will perform as well or nearly as well as portfolios with slicing and dicing in every which way. I'm now rethinking my decision to go with the Coffeehouse portfolio and will just go to the 3/4 fund portfolio and forget about worrying if value stocks are getting hammered or not.
Thanks Simba for your significant efforts on this.
Retired |
Two-time in top-10 in Bogleheads S&P500 contest; 18-time loser
I have the updated the spreadsheet with the 2010 returns. As Cb said i am waiting for a couple more funds to announce the returns in addition to the CPI numbers to be announced. Once I have them, I'll post the latest revision.
Mike/Don (DP) - the latest version fixes the issues you raised. Thanks for bringing them up.
I think the formula in P67 should be =AVERAGE(P$118:P$143)
I have search this forum many times trying to find another typo reported about this spreadsheet... I can not find it. I think the typo was if year 1 had a negative return? Do you see this typo?
Mike
As Merton Miller, a Nobel laureate at the University of Chicago, puts it, "I'll never understand why they call bonds 'fixed' income."
Simba - much anticipated. Your spreadsheet is an absolute must-have in my investment toolchest. Thanks for all you've done.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs
Just found this spreadsheet. It is amazing what has been put together, but the Fundadvice is slightly off from what is currently put into the Lazy Portfolio sheet. to be accurate, it looks like (unfortunately 2 more columns of data needed to be added:
International Small Cap
Intl Small Cap Value
See: Fundadvice dot com Ultimate Buy and Hold Strategy
Portfolio 6.
Anyway of adding those 2 columns since the update for 2010 is probably being worked on as we speak?
From Bill Bernstein's book, The Four Pillars of Investing (2010)
The study of financial history is an essential part of every investor's education. It is not possible to precisely predict the future, but knowledge of the past often allows us to identify financial risk in the here and now. Returns are uncertain. But risks, at least, can be controlled.
Hey I'm a little bit confused. I made a hypothetical portfolio that is 100% t-bills/treasury money market and I'm showing a 2.76% percent return for 2010.
That's not right, is it?
Thanks for your wonderful tool. It has really helped me analyze my portfolio