Portfolio Allocation Timing Models

Ask your investment, budget, and other money related questions here
Seppia
Posts: 1057
Joined: Tue Aug 30, 2016 9:34 am
Location: Italy

Re: Portfolio Allocation Timing Models

Post by Seppia » Sun Jun 23, 2019 7:54 pm

I always forget that in civilized countries people have access to all trading instruments in tax advantaged accounts.
Here in Italy a strategy such as yours would be only doable in regular brokerage accounts, and any extra return would probably be more than compensated by the huge tax hit of trading in and out monthly.

frihet
Posts: 190
Joined: Thu Jun 06, 2013 10:33 pm
Contact:

Re: Portfolio Allocation Timing Models

Post by frihet » Sun Jun 23, 2019 8:00 pm

This is not my analysis. But it is my understanding that PV doesn't take friction costs into account . So these numbers are not real in that sense......

With ETFs there are spreads which can add up considerably and with funds there are time out of the market while changing. Then there is of course taxes. You obviously want to trade this in none taxable accounts. Which we have in Sweden( small fixed account tax yearly, not on every transaction)

I personally like educated gambles......all investing is an educated gamble in my mind. But diversify between your gambles and do not bet everything on one horse/stock/strategy/asset class

Nomad
Posts: 182
Joined: Wed May 16, 2018 5:23 pm
Location: UK

Re: Portfolio Allocation Timing Models

Post by Nomad » Sun Jun 23, 2019 8:04 pm

bigato wrote:
Sun Jun 23, 2019 7:30 pm
Well, one thing that you could do is to try and tweak the traditional portfolio allocation strategies for that same short period of years and see if you can come up with something that has similar returns. You probably can, and that just means that if you squeeze a small sample hard enough, you can get pretty much any results you want from it. I personally would only consider it with a much bigger historic data. Other than that you're gambling, which is fine if that's what you want.
I would say it is the opposite of gambling.
Putting money into an asset class that has been consistently losing value for a period of six or twelve months and hoping that this month it will change direction is gambling.
What normally happens is that trends continue for quite a long time before stopping, wobbling and then sometimes going in the opposite direction for a while.
My strategy at the moment is having money in diversified asset classes that are all making money.

frihet
Posts: 190
Joined: Thu Jun 06, 2013 10:33 pm
Contact:

Re: Portfolio Allocation Timing Models

Post by frihet » Tue Jun 25, 2019 7:12 am

https://allocatesmartly.com/aggressive- ... llocation/

Here is a backtest of a similar strategy from Meb Faber with more accurate data. Still good but not as good as the original testing done by Meb.

"Less optimistic results:

These results, while very good, are considerably less optimistic than those presented in the original paper. That’s partially a result of the addition of transaction costs, but even if we assumed costs to be zero, our annualized returns would still fall short by 2.6% (Agg 3) and 3.7% (Agg 6) during overlapping years.

The difference is largely the result of using more accurate proxies to simulate ETF data prior to each ETF’s launch. The original paper used Fama-French data to represent equity asset classes which, as shown here, often provides an overly-optimistic view of asset class returns. When running the test without transaction costs and with Fama-French data, our returns are similar to those in the original paper."

Personally though I have realized that I need more stock exposure to have a chance for FI long term and I find these kind of strategies a good compromise for part of the portfolio. Having been in Permanent Portfolio/Golden Butterfly type of portfolios since the financial crises basically, I''ve "lost" a lot gains, if I had gone with dividend investing i would have been FI many years ago with a runaway portfolio......I just can't make myself being invested in stocks and risk the kind of drawdowns I experienced 10 years ago. Especially not now with overvalued markets, this strategy's semi active approach at least gives a historical possibility of smalller drawdowns. While still taking part in the upside. Like I said I would not put all my money into it and I don't. But certainly some. What I would also like to add is a quant value approach. As from what I've read value and momentum/trend following are often uncorrelated. But these strategies have large drawdowns so have to think about it some more....

bigato
Posts: 2117
Joined: Sat Mar 05, 2011 12:43 pm

Re: Portfolio Allocation Timing Models

Post by bigato » Tue Jun 25, 2019 7:14 am

Nomad: So you just conveniently ignore that the traditional portfolios works in backtesting going back to the 1930’s and classify it as a gamble? What do you call a strategy that you have only tested for a recent 13 years period?

bigato
Posts: 2117
Joined: Sat Mar 05, 2011 12:43 pm

Re: Portfolio Allocation Timing Models

Post by bigato » Tue Jun 25, 2019 7:22 am

Frihet: the main difference in the strategy is that the timing assumes that momentum is a thing most of the time and that you should be able to take advantage of it tweaking some triggers, right? So the best way to find out how good it is would be to test it with asset classes for which we have more historic data. Like, how well does it work if you backtest the Permanent Portfolio using the timing strategies going back to the beginning of last century? Is is better or worse than the traditional portfolio rebalancing strategy?

If it only works with assets created recently, chances are that the reasons for the numbers are not in the timing strategy but in the assets themselves. And that is likely to change. We need testing for longer periods of time, doing it with only recent data won’t help.

Seppia
Posts: 1057
Joined: Tue Aug 30, 2016 9:34 am
Location: Italy

Re: Portfolio Allocation Timing Models

Post by Seppia » Tue Jun 25, 2019 9:40 am

frihet wrote:
Tue Jun 25, 2019 7:12 am
Personally though I have realized that I need more stock exposure to have a chance for FI long term and I find these kind of strategies a good compromise for part of the portfolio. Having been in Permanent Portfolio/Golden Butterfly type of portfolios since the financial crises basically, I''ve "lost" a lot gains, if I had gone with dividend investing i would have been FI many years ago with a runaway portfolio......
This part here to me is a giant red flag.

Seppia
Posts: 1057
Joined: Tue Aug 30, 2016 9:34 am
Location: Italy

Re: Portfolio Allocation Timing Models

Post by Seppia » Tue Jun 25, 2019 9:45 am

Nomad wrote:
Sun Jun 23, 2019 8:04 pm
I would say it is the opposite of gambling.
Putting money into an asset class that has been consistently losing value for a period of six or twelve months and hoping that this month it will change direction is gambling.
What normally happens is that trends continue for quite a long time before stopping, wobbling and then sometimes going in the opposite direction for a while.
My strategy at the moment is having money in diversified asset classes that are all making money.
Again it depends on what your time horizon is.
Assuming stocks will continue in the same direction usually works... until it doesn’t, and when it doesn’t, it usually doesn’t A LOT.
7 of the 10 most valuable companies in the world are tech companies, and tech is by far the sector that performed the best in the last decade.
Look back a few years and they were all railroads.
Look back less years and they were all energy companies.

Would you bet on tech?
For the next three months, maybe.
For the next three decades, not at all.

frihet
Posts: 190
Joined: Thu Jun 06, 2013 10:33 pm
Contact:

Re: Portfolio Allocation Timing Models

Post by frihet » Tue Jun 25, 2019 1:32 pm

Seppia I understand why you call it a red flag. But i don't see it that way as i am risk averse with most of my portfolio. I consider that part as an air bag that will give me plenty of time to find work/income/cut expenses. This part of the portfolio is enough to sustain me, bare bones FI style.

But I would prefer to be able to spend more. For that to be possible I need to take on more risk with a calculated part or work longer. Work longer is simply not an option for me anymore. I want to be free while I'm still relatively young. i don't have responsibility for anyone but myself so i can take that risk, my understanding is that you are super risk averse and have a family to support? Each to his own I guess.

Congratulations for having the gutso to be in stocks by the way :) i f I had listened less to Macrovoices and real vison and the likes. I might have had as well. To let the trend be your friend is not such a bad idea after all...

One more thing for arguments sake. Lets say you follow a momentum strategy like this. In your list you include a tech fund, energy and railroads among with other uncorrelated things, gold, bonds, and so on. If tech will be the in or out of favour for the coming decade is simply not an issue as you are only choosing the 3 best performing ones each month. The obvious weak spot is of course whip saws that no real trends develope and you switch funds all the time. But that is a risk I'm willing to take, with some of my money.

Nomad
Posts: 182
Joined: Wed May 16, 2018 5:23 pm
Location: UK

Re: Portfolio Allocation Timing Models

Post by Nomad » Tue Jun 25, 2019 2:10 pm

bigato wrote:
Tue Jun 25, 2019 7:14 am
Nomad: So you just conveniently ignore that the traditional portfolios works in backtesting going back to the 1930’s and classify it as a gamble? What do you call a strategy that you have only tested for a recent 13 years period?
Although traditional fixed portfolios work, so does putting money in a mattress - it just doesn't work very well in comparison.
None of the online analysis websites I have seen contain data going back to the 1930's, 1985 is the about the earliest but this depends on
which asset classes/tickers you select.
Where traditional portfolios fare very badly compared to a dynamic allocation approach is in the drawdown stage.
You should try out some scenarios. If you have historical data available you can backtest with this.
Last edited by Nomad on Tue Jun 25, 2019 2:19 pm, edited 1 time in total.

Nomad
Posts: 182
Joined: Wed May 16, 2018 5:23 pm
Location: UK

Re: Portfolio Allocation Timing Models

Post by Nomad » Tue Jun 25, 2019 2:13 pm

Seppia wrote:
Tue Jun 25, 2019 9:45 am
Again it depends on what your time horizon is.
Assuming stocks will continue in the same direction usually works... until it doesn’t, and when it doesn’t, it usually doesn’t A LOT.
7 of the 10 most valuable companies in the world are tech companies, and tech is by far the sector that performed the best in the last decade.
Look back a few years and they were all railroads.
Look back less years and they were all energy companies.
Exactly, when a stock market crash happens, there can be a short sharp shock - but then there are about 18 months of misery as prices
slowly dwindle. A dynamic allocation model approach eliminates most of the misery.

Nomad
Posts: 182
Joined: Wed May 16, 2018 5:23 pm
Location: UK

Re: Portfolio Allocation Timing Models

Post by Nomad » Mon Jul 01, 2019 4:16 pm

Ok, the funds I picked using a market timing strategy last month were these.

Fundsmith Equity 25% (440.97)
First State Greater China Growth 10% (910.27)
Lindsell Train Japanese Equity 10% (272.89)
Fidelity India Focus 5% (274.4)
Fidelity Global Property 15% (200.3)
First State Global Listed Infrastructure 10% (310.55)
LF Lindsell Train UK Equity 10% (186.59)
Jupiter European 5% (2597.56)
Xtrackers Physical Gold ETC 10% (10266.39)

I've liquidated the Indian fund and the property fund. and sold the two Lindsell Train funds to buy Lindsell Train Global Equity.
Numbers in brackets are todays prices.

Fundsmith Equity 28% (466.74) up 5.84%
First State Greater China Growth 15% (988.42) up 8.59%
First State Global Listed Infrastructure 8% (318.65) up 2.61%
Jupiter European 8% (2723.28) up 4.84%
Lindsell Train Global Equity 25% (267.20)
Xtrackers Physical Gold ETC 14% (10731.19) up 4.65%

Overall I think I'm up 3% for the month.
This is pension #2 88744.
The experiment continues.

User avatar
jennypenny
Posts: 6240
Joined: Sun Jul 03, 2011 2:20 pm
Location: Stepford USA

Re: Portfolio Allocation Timing Models

Post by jennypenny » Mon Jul 29, 2019 5:32 am

Strategic Rebalancing paper (PDF)

Concluding Remarks:
A pure buy-and-hold portfolio is untenable for most investors as it leads to a highly concentrated, undiversified portfolio. However, a 60-40 stock-bond portfolio (our use case) that rebalances every month to the 60:40 target ratio loses several percentage points more than a buy-and-hold portfolio during periods of continued stock market drawdowns. In essence, rebalancing to a constant asset mix means selling winners and buying losers, which hurts at times when the stock performance (relative to that of bonds) is trending. We show that the concavity induced by rebalancing is effectively countered with a trend exposure, which exhibits convexity and can be either implemented as a direct allocation to a trend strategy or with a strategic trend-based rebalancing rule.

While our focus is on countering the concavity induced by rebalancing, other considerations matter in practice as well. For example, investors can also use monthly in- and out-flows to move back toward the target asset mix. For example, Chambers, Dimson, and Ilmanen (2012) mention that the Norwegian Government Pension Fund Global directs monthly inflows into the asset class that is most underweight relative to the benchmark. For taxable investors, rebalancing using income has the added benefit that no assets need to be sold, which can be tax efficient; see Colleen, Kinniry, and Zilbering (2010).

Finally, we note that a stock-bond trend exposure is just one way to mitigate drawdowns at times of continued stock market losses. An investor has more arrows in her quiver. A good starting point is a more diversified portfolio that includes more asset classes and has an international exposure. An allocation to a broader a broader trend strategy that benefits from trends in other macro assets at times of equity market distress may further dampen equity market losses, see Hamill, Rattray, and Van Hemert (2016). And Harvey et al. (2018) study volatility targeting and show that it can help manage the risk of a 60-40 stock-bond portfolio.

Nomad
Posts: 182
Joined: Wed May 16, 2018 5:23 pm
Location: UK

Re: Portfolio Allocation Timing Models

Post by Nomad » Tue Jul 30, 2019 5:26 pm

jennypenny wrote:
Mon Jul 29, 2019 5:32 am

Finally, we note that a stock-bond trend exposure is just one way to mitigate drawdowns at times of continued stock market losses. An investor has more arrows in her quiver. A good starting point is a more diversified portfolio that includes more asset classes and has an international exposure. An allocation to a broader a broader trend strategy that benefits from trends in other macro assets at times of equity market distress may further dampen equity market losses, see Hamill, Rattray, and Van Hemert (2016). And Harvey et al. (2018) study volatility targeting and show that it can help manage the risk of a 60-40 stock-bond portfolio.
[/i]
From experimenting with the Timing Strategies on the page https://www.portfoliovisualizer.com/
Draw downs during the accumulation stage are not disasterous unless you stop accumulating. If you carry, you are buying assets at a discount and
you will soon recover from the short fall the minute the market starts to recover.
Draw downs in a fixed portfolio during the 'draw down'/spending stage can be catastrophic but using a timing/reactive mode that gets you out of asset
classes that are in decline is the only strategy I've seen that stops the rot.

Nomad
Posts: 182
Joined: Wed May 16, 2018 5:23 pm
Location: UK

Re: Portfolio Allocation Timing Models

Post by Nomad » Tue Jul 30, 2019 5:50 pm

Ok, I've done my analysis... This month will be largely gold, mainly US stock and emerging markets. Will be essentially keeping the same funds.

So from 1st July until 30th July prices have increased.
Fundsmith Equity (466.74) to (491.51) up 5.31%
First State Greater China Growth (988.42) to (1052.39) up 6.47%
First State Global Listed Infrastructure (318.65) to (332.14) up 4.23%
Jupiter European (2723.28) to (2785.91) up 2.3%
Lindsell Train Global Equity (267.20) to (280.42) up 4.95%

Note, I've switched the Gold ETF due to Deutche Bank problems...
Gold Bullion Securities ETF new price (11037) I think the value of the Gold ETF was nonetheless up a couple of percent.

So, up approximately up 4.35% for the month which is pretty good.
Possibly I should sell the Jupiter European as it wasn't recommended by analysis and is the lowest performing...
So far, I must admit I'm pretty blown away by the performance of the market time strategy.

Nomad
Posts: 182
Joined: Wed May 16, 2018 5:23 pm
Location: UK

Re: Portfolio Allocation Timing Models

Post by Nomad » Sat Sep 07, 2019 6:18 am

Slightly late analysis this time. This time dropping the European fund was recommended and buying more Gold.

From 30th July until 7th September.
Gold Bullion Securities ETF new price (11037) to (11590) up 5.01%
Fundsmith Equity (491.51) to (485.08) down -1.31%
First State Greater China Growth (1052.39) to (1028.58) down -2.26%
First State Global Listed Infrastructure (332.14) to (334.97) up 0.85%
Lindsell Train Global Equity (280.42) to (280.48) up 0.02%

So overall, up 0.46% for the period.
This pension is up to £92551 from £88744 on 1st July.

Post Reply