True paralysis analysis

[hotwired]

Good morning,
I feel like a real "newbie" today. I've been investing for a long time. I'm 46 years old, FI, and I love learning about investing, strategizing, etc. but I find myself getting overwhelmed, letting my portfolio get over complicated and not really "owning" it. I have good reasons for buying what I buy but I don't necessarily follow up, and I have too many things to make rebalancing simple. In short, my ambition on the front end outweighs my ambition on the back end. (I've got a big house but I'm too lazy to keep it clean!!)

So I'm in teh process of simplifiying. I looked at my vanguard account with 12 funds, and then my wife and daughter's with 2-3 funds each (STAR, total bond and div grwth for daughter, Total market, div growth, total bond for wife) and they both have outperformed my "really smart" portfolio over 1,3,5, and 10 years. (mine is divided up into small cap, int'l RE, us RE, etc. etc. and is overdiversified, I even have some wellington in there)

Where I'm paralyzed is this: I really want to start all over. I have a ROTH and SIMPLE and I just want to start over and make a nice simple AA and rebalancing strategy that I will follow through on. I'm realizing slowly that just having a sensible strategy, no frills is fine. There's always a book or article that leads me astray ((... if I just follow x,y, z, backtesting shows I'd have an extra gazillion by now!!)) Problem is, I'm too impatient. My personality thrives on simple "ownable undertsandable" portfolio that I can get carried away putting money into.

I do like the IVY portfolio strategy. MY ROTH at Fidelity is what I use for that. I have commodity ETF, ARBFX, JOF, EWJ, PSP, LINE, DBU, VNM, AGNC, RJI, RVT, SIRI, etc. alot of international stock and bond, alot of value, MLP's, and one actual stock (SIRI). I can probably handle keeping this ROTH mostly the way it is and call it my "alternative" portfolio. I know enough to stay diversified and keep a fairly equal weighting but I'd like to get my larger Vanguard portfolio simplified beyond recognition. I'd like to keep RE - both int'l and US but otherwise keep it simple. 60/40-ish stock bond.

Any feedback or suggestions appreciated. Iknow I haven't asked a black and white crystal clear question. I've just laid my cards on the table a bit and would like some feedback, direction, etc.

Thanks SO much!!

[hotwired]

For my Vanguard SIMPLE: I'm beginning to have some answers form in my mind as I ask questions. I do have SOME assumptions that I want to stick to: over time (20 years or more) value and small company stock will win, hence I will "tilt" weightings int hat direction. I assume (but don't know) that this holds true for international stock as well.

Being a RE investor (49 apartments), I value cash flow so I want to keep the div growth fund. I also want to keep durations short on the bond fund side. I can't tell when, but can it be much more time b4 rates rise I won't worry about high yield on the bond side as my ROTH takes care of that with various int'l bond funds and MLPs, etc. (even floating rate fund)

that being said, I'm considering revamping my VANGUARD SIMPLE thusly (40-40-20 stock bond, RE but individual weightings within the asset classes will be equal)

Total stock mkt
div grwth
small cap value

emerging markets
int'l small cap (ex-us)
int'l value

gnma
tips
st inv grade

ex us REIT
us REIT

[LOL!]

One can get lots of portfolio help (i.e. more eyeballs on your predicament) at Bogleheads Investing Advice and Info

[pb4uski]

Forget accounts for now. Think in terms of what your personal target AA is across all you accounts, how your current AA across all accounts lines up with your target AA and then make changes to rebalance your overall AA to target. One way to do it would be to plug all your investments into VG and then look at Portfolio Watch, select a AA consistent with your risk appetite and see what Portfolio Watch recommends.

For example, my target is 60% stock (45% domestic equities, 15% international equities) and 40% fixed income. Given the current low interest rate environment, within fixed income I prefer low to moderate interest sensitive bond funds and a mix of investment grade corporate and high yield corporates.

Once you know your target AA and tickers, then you can "allocate" investments to accounts (taxable, tax-deferred, tax-free). I put all my fixed income in tax deferred and stocks in taxable and tax-free.

I also use Portfolio Tester to compare attributes of my before and after portfolios. Another alternative is to go through VG's financial planning process and they will make recommendations on what changes you should make or Financial Engines and see what changes they recommend.

Of course, if you are making changes to taxable accounts you'll want to keep tax implications in mind.

[jerome len]

If you have enough with Vanguard they'll do a free analysis of what you have, ask you your goals and then give you their recommondations. If you want to keep it simple, buy total stock, intermediate bond fund and a life cycle fund, based on your year of planned retirement. I'm like you, I have a dozen and I think it's good....it's diversified. I'm glad I have REITs and emerging markets since they have done well. But.....listen to all......there are a lof of ways to invest......you have to do what helps you sleep well at night. good luck!!!!!

[easysurfer]

To make things simple, I'd go with following the indexes.

More specifically, Total US Stock, Total International Stock, Total Bond.

From there it's just a matter of rebalancing to your target allocations when your actual allocations drift.

Plus, by just following the indexes, you don't have to worry whether you have a good fund manager, etc.

[Lsbcal]

1. Select your AA, stocks/bonds/cash

2. Select a set of benchmarks, for instance:
.....Wellington (AA 65/35)
.....Wellesley (AA 50/50)
.....Total Stock Mkt/ Total International/Total bond mkt (set to your AA from #1)
.....Your guess at best simple portfolio for you (maybe a few of these)

3. Backtest to see how they have done in different market environments. Yearly data should be pretty easy to come by. See Vanguard, Morningstar.

4. Follow these for maybe a few quarters. See how you feel about each of these on a day to day, month to month basis.

5. Select one of these as your long term choice.

BTW, I'd get rid of the REITs and other stuff. REITs are already in your broad funds. Just focus on the broad asset classes. Don't treat the Roth differently as it's just as important and tax free too -- it is not mad money.

[hotwired]

thank you Lsbcal,
that is very simple! BTW, what do you like for backtesting? I want to really dig in and see how much some of these choices affect total return. (i.e. etfreplay, quantext, etc.)

[Lsbcal]

Quote:

Originally Posted by hotwired thank you Lsbcal, that is very simple! BTW, what do you like for backtesting? I want to really dig in and see how much some of these choices affect total return. (i.e. etfreplay, quantext, etc.)

I'm not quite sure I'm interpreting your question correctly. When I backtest I generally use monthly data from Yahoo, i.e. their adjusted close prices. Monthly because I don't want any strategy that I have to follow day to day -- life can get in the way. I download into a spreadsheet and crank away. Ultimately I'm shooting for backtesting back to the 1929 crash. So I've even used French Fama data to simulate stuff backwards. This is way more trouble then most people have a tolerance for. But it's fun for me and satisfies an urge to analyze everything.

Backtesting ETF's is a problem because they don't go back many years. So I generally use fund data and Yahoo can go back maybe 20 years in some cases. Then there is MSCI data and French-Fama.

It can take a lot of time to do this stuff correctly and I recommend letting your analysis "simmer" for quite awhile. That is why I mentioned following things over several months or even a few years. You will find out more about yourself and your risk tolerances.

[hotwired]

I'm curious, do you do it all by hand? I was thinking of some of the "monte carlo" simulators and "portfolio testers". there's a fairly expensive one ($150 +/-) called Quantext Portfolio Planner which is an excel spreadsheet but will work with nearly any stock symbol. Ben Stein uses it so it can't be too bad! I have it here, but need to renew it.

thanks again for your thorough and helpful replies

[Lsbcal]

I want it under my control so it's all by hand. Pretty generic Excel with a little bit of Visual Basic. I'm hoping that this will be usable for years. Don't want to rely on questionable software cycles and other businesses (except for Microsoft Excel).

Added note: For most backtesting I'd think that the available free data + tools satisfies most investor needs. My comments today on backtesting were in regards to how I personally do it. However, my backtesting is for research purposes that go further then most people want to go and the techniques would probably horrify a few staunch buy-hold types. So you could safely ignore some of my comments ... maybe all of them if you want to.

[pb4uski]

Why not look at blended returns for 1, 3, 5 and 10 years?

[target2019]

Look for Trev backtest spreadsheet and EZBacktest. These are two tools worth keeping. The first has historical data loaded to spreadsheet. You can create new port of Vanguard funds and test against benchmarks. The second pulls live data from yahoo and compares your portfolio against one benchmark. The software has other features, so check the options.
If a simple portfolio has outperformed yours, you are probably light on small and mid caps, as well as emerging markets. That is looking back, though. Put your port into EZB, and tweak a fund or two to see what I mean.

[mistershankly]

Quote:

Originally Posted by Lsbcal I download into a spreadsheet and crank away. Ultimately I'm shooting for backtesting back to the 1929 crash. So I've even used French Fama data to simulate stuff backwards

Lsbcal, what portion of your analysis of such a long range do you use to make decisions in the present time? Researching back to 1929 may cover some atypical times in the market (e.g. less volatility, less complexity and transaction speeds affecting the market, etc) that may not be reflective of the recent 1-2 decades. Do you look for cyclical trends that are inherent in any economic system regardless of the socioeconomic climate at the time or is your interest beyond a certain decade mostly for analytical exercise? BTW, I do as you do and crank out my own spreadsheets rather than rely on others' software... I'm an analysis nerd like that and it sure beats watching TV at times

[Lsbcal]

Quote:

Originally Posted by mistershankly Lsbcal, what portion of your analysis of such a long range do you use to make decisions in the present time? Researching back to 1929 may cover some atypical times in the market (e.g. less volatility, less complexity and transaction speeds affecting the market, etc) that may not be reflective of the recent 1-2 decades. Do you look for cyclical trends that are inherent in any economic system regardless of the socioeconomic climate at the time or is your interest beyond a certain decade mostly for analytical exercise? ...

First before I get flamed, I'm not recommending my approach as a general one. I have the data from recent decades in very precise form and some reasonable semblance of it going back to the 1929 crash. I'm looking for a rough rule when to pull back my equity allocation, perhaps to zero. I'm not trying to do day to day trading. Maybe one signal every 4 years or so, quite irregular too. It is "almost" buy-hold.

Data sources come from:
1) Yahoo historical data
2) MSCI data (Vanguard uses MSCI indexes)
3) Federal reserve interest rates
4) Schiller PE10 data, his spreadsheet updated monthly

I use the data going way back to do an out of sample test.

To be very frank, I don't think it is possible to avoid a data mining criticism. If something works in recent decades, and if it doesn't work in the distant past, then one just dumps the idea or modifies it. But I think we all are a bit of data miners, even buy-hold investors rely on historical evidence.

[samclem]

hotwired,
If I may, some observations:
- You are too "wonky" to accept a radical simplification of your investments. It's evident in the investment basket you've already created and in the questions you're asking about backtesting. There's nothing wrong with this.
- You recognize that you don't do a good job of following through. And part of the reason may be the shear volume of the types of investments you have (see previous paragraph), the newsletters/gurus/systems you are monitoring, the backtesting experiments you want to try, etc. It's too much and you probably feel you should be doing it all the time, that you are missing out on possible gains.

So, instead of shooting to simplify your portfolio, why not shoot to simplify the maintenance of it? First design a desired AA (with target ranges for each asset category) you can pick any niche categories you want, but write down in a notebook exactly why you are making each investment (what's its role in your portfolio). Then align your investments with the plan. Here's what you do different: Don't look at it or mess with it except for once a year (in conjunction with tax time if that suits you). Then do a mechanical rebalance to get everything set again. This is a maintenance schedule you can stick to, it will prevent you from panic buying/selling, it will remove the guilty feeling that you should "be doing something" for 11 months of the year, and you'll probably end up ahead of the game.

If you find yourself getting eager to "do something", you can always set up a small hobby portfolio to allow for trading and following up on the latest "hot sector" financial pornography. That'll keep you occupied while the bulk of your holdings grow unmolested.

[REWahoo]

Quote:

Originally Posted by samclem That'll keep you occupied while the bulk of your holdings grow unmolested.

"Portfolio molestation"

Not a crime, but the financial penalties can be severe.