Page 1 of 1

Market timing using Gold-Lumber indicator

Posted: Sun Sep 10, 2017 5:56 am
by Lauretta
I came across this paper yesterday. https://papers.ssrn.com/sol3/papers.cfm ... id=2604248
Apart from the apparent(?) spelling mistake in the title it looks like it's good stuff - one of the things that looks interesting is that compared to say momentum or moving averages rules which I was told can't really be applied to small caps because they're too volatile, this method seems to work well for small caps too.
Has anyone been looking into it? I'm not sure e.g. what the turnover would be and whether after costs this strategy is still worth considering.
I must say I have only recently begun to learn about investing - I am reading about all these methods but in practice I stick to a B&H strategy also because I think it fits my temperament better. But it's good to learn about other methods. Perhaps also to put a fraction of one's wealth in these strategies to diversify further ;-)

Re: Market timing using Gold-Lumber indicator

Posted: Sun Sep 10, 2017 7:02 am
by Chip
I skimmed the paper. Honestly it seems like a lot of curve fitting to me.

I was amazed to see them quote wikinvest for the statement (p. 3) that:

"It is estimated that “one-third of the forestland in the United States is publicly owned and has been withdrawn from production…[and] of the remaining 500,000 acres [emphasis added], 29% is publicly owned and contributes very little to the Nation’s timber output.”

So the nation's total "forestland" is only 750,000 acres? They missed the mark by a factor of 1000. And "wikinvest" as a source?

I wonder why they chose November, 1986 as their start date. I think both gold and lumber futures were available back to the mid-70s.

Re: Market timing using Gold-Lumber indicator

Posted: Sun Sep 10, 2017 7:25 am
by Lauretta
Chip wrote: Sun Sep 10, 2017 7:02 am
So the nation's total "forestland" is only 750,000 acres? They missed the mark by a factor of 1000. And "wikinvest" as a source?
Ah ok. Thanks for sharing your input :) Then it looks like the errors go beyond the spelling of 'it's' in the title ;-)

Re: Market timing using Gold-Lumber indicator

Posted: Mon Sep 11, 2017 10:55 pm
by Phineas J. Whoopee
Across the US economy and population, increased spending on admission to professional sporting events is was strongly and inversely correlated with per capita consumption of high-fructose corn syrup.
PJW

Re: Market timing using Gold-Lumber indicator

Posted: Tue Sep 12, 2017 5:54 am
by Lauretta
Phineas J. Whoopee wrote: Mon Sep 11, 2017 10:55 pm Across the US economy and population, increased spending on admission to professional sporting events is was strongly and inversely correlated with per capita consumption of high-fructose corn syrup.
PJW
:D :D

Re: Market timing using Gold-Lumber indicator

Posted: Tue Sep 12, 2017 10:12 pm
by aegis965
Lauretta wrote: Sun Sep 10, 2017 5:56 amcompared to say momentum or moving averages rules which I was told can't really be applied to small caps because they're too volatile
By small caps do you mean small caps as an asset class or individual stocks? If it's the former, I don't see how momentum or MA won't work. If it's the latter, adjusted slope could help:
Andreas Clenow wrote:First, we use exponential regression to measure the momentum. As opposed to its linear cousin, we get the slope expressed in percent. The daily slope gives you a number with many decimals that's hard to relate to, so let's annualized that sucker. Now you have a number which answers the question "How many percent would this stock make in a year, if it was to continue the same trajectory as the recent past?". No, we don't expect that to happen, but it gives us a number that we can relate to.

But what about volatility, you ask. Wouldn't this reward extreme situations, like takeover bids and crazy vola? Yes, but that's where our friend R2 comes into play. The R2, of coefficient of determination, is a number between 0 and 1 which tells us how well our regression actually fits the data. It would be 1 if all the actual data points are exactly on the line.

So, now we simply punish the volatile stocks by multiplying all of our annualized regression slopes by the R2. A stock with a nice gradual slope will have a high number and won't take much of a hit. An overly volatile stock filled with big gaps will get pushed down the ladder.

Now we simply buy stocks from the top. Positions are vola parity and rebalancing done monthly. Voila. Simple and robust.
BTW This ETF utilizes momentum signals to switch between high beta portfolio and low vol portfolio. Pretty cool idea and IMO more robust than the gold-lumber thingy.

Re: Market timing using Gold-Lumber indicator

Posted: Tue Sep 12, 2017 11:09 pm
by pkcrafter
Lauretta wrote: Sun Sep 10, 2017 5:56 am I came across this paper yesterday. https://papers.ssrn.com/sol3/papers.cfm ... id=2604248
Apart from the apparent(?) spelling mistake in the title it looks like it's good stuff - one of the things that looks interesting is that compared to say momentum or moving averages rules which I was told can't really be applied to small caps because they're too volatile, this method seems to work well for small caps too.
Has anyone been looking into it? I'm not sure e.g. what the turnover would be and whether after costs this strategy is still worth considering.
I must say I have only recently begun to learn about investing - I am reading about all these methods but in practice I stick to a B&H strategy also because I think it fits my temperament better.

That's good, welcome to the club. :happy

But it's good to learn about other methods. Perhaps also to put a fraction of one's wealth in these strategies to diversify further ;-)

Nope. :happy

Paul

Re: Market timing using Gold-Lumber indicator

Posted: Tue Sep 12, 2017 11:25 pm
by BolderBoy
Phineas J. Whoopee wrote: Mon Sep 11, 2017 10:55 pm Across the US economy and population, increased spending on admission to professional sporting events is was strongly and inversely correlated with per capita consumption of high-fructose corn syrup.
PJW
Personally, all my money is in Frozen Concentrated Orange Juice futures. (with apologies to the movie, "Trading Places")

Re: Market timing using Gold-Lumber indicator

Posted: Tue Sep 12, 2017 11:41 pm
by Uncle Pennybags
That is a "fake" scholarly paper site. Anyone can submit anything and they will publish it. The "it's" in the title was placed by the web site. Its spelled correctly in the scholarly paper. Shocking fake stuff on the Interweb.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 12:01 am
by aegis965
SSRN is fake!?

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 3:57 am
by Lauretta
aegis965 wrote: Tue Sep 12, 2017 10:12 pm
Lauretta wrote: Sun Sep 10, 2017 5:56 amcompared to say momentum or moving averages rules which I was told can't really be applied to small caps because they're too volatile
By small caps do you mean small caps as an asset class or individual stocks? If it's the former, I don't see how momentum or MA won't work. If it's the latter, adjusted slope could help:
Andreas Clenow wrote:First, we use exponential regression to measure the momentum. As opposed to its linear cousin, we get the slope expressed in percent. The daily slope gives you a number with many decimals that's hard to relate to, so let's annualized that sucker. Now you have a number which answers the question "How many percent would this stock make in a year, if it was to continue the same trajectory as the recent past?". No, we don't expect that to happen, but it gives us a number that we can relate to.

But what about volatility, you ask. Wouldn't this reward extreme situations, like takeover bids and crazy vola? Yes, but that's where our friend R2 comes into play. The R2, of coefficient of determination, is a number between 0 and 1 which tells us how well our regression actually fits the data. It would be 1 if all the actual data points are exactly on the line.

So, now we simply punish the volatile stocks by multiplying all of our annualized regression slopes by the R2. A stock with a nice gradual slope will have a high number and won't take much of a hit. An overly volatile stock filled with big gaps will get pushed down the ladder.

Now we simply buy stocks from the top. Positions are vola parity and rebalancing done monthly. Voila. Simple and robust.
BTW This ETF utilizes momentum signals to switch between high beta portfolio and low vol portfolio. Pretty cool idea and IMO more robust than the gold-lumber thingy.
Thanks! :) I'll look at this in detail. The results on the webpage make sense: protect you in bear markets at the cost of underperformance in bull markets. I had briefly seen a paper by Philippe Bouchaud, a very smart French physicist who became a quant, where he used exponential moving averages.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 7:24 am
by Uncle Pennybags
aegis965 wrote: Wed Sep 13, 2017 12:01 am SSRN is fake!?
It is not peer reviewed.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 7:38 am
by petulant
Thanks for publicizing OP. Now my outsized returns will be reduced by the rush of investors copying the strategy. Ugh.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 8:01 am
by Uncle Pennybags
Oil used to be the leading indicator. It took the market six months to realize that fracking drove down the price not a lack of demand. This is why a retail investors' only hope is the Boglehead way of investing.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 8:26 am
by Lauretta
petulant wrote: Wed Sep 13, 2017 7:38 am Thanks for publicizing OP. Now my outsized returns will be reduced by the rush of investors copying the strategy. Ugh.
good joke :thumbsup

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 8:32 am
by aegis965
Lauretta wrote: Wed Sep 13, 2017 3:57 amI had briefly seen a paper by Philippe Bouchaud, a very smart French physicist who became a quant, where he used exponential moving averages.
Wow, this guy is doing some interesting stuff. Thanks. I'm reading the tail risk premia paper, very rewarding so far.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 8:40 am
by Lauretta
aegis965 wrote: Wed Sep 13, 2017 8:32 am
Lauretta wrote: Wed Sep 13, 2017 3:57 amI had briefly seen a paper by Philippe Bouchaud, a very smart French physicist who became a quant, where he used exponential moving averages.
Wow, this guy is doing some interesting stuff. Thanks. I'm reading the tail risk premia paper, very rewarding so far.
yes and he has done very cool stuff in physics too :idea: ... He's very clever I think. And he seems to have done very well :moneybag :moneybag - a friend of mine knows him well and he told me he bought a theatre in Paris for his wife (who's also a scientist but she's interested in acting...)

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 8:43 am
by midareff
Uncle Pennybags wrote: Wed Sep 13, 2017 8:01 am This is why a retail investors' only hope is the Boglehead way of investing.
Quote from Uncle Pennybags edited for the purpose of this post.

As the hedge fund guru who just paid up his bet, a million bucks to Buffet's designated charity, has found that holding the S&P500 at low cost is very powerful medicine.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 10:10 am
by David Jay
Phineas J. Whoopee wrote: Mon Sep 11, 2017 10:55 pm Across the US economy and population, increased spending on admission to professional sporting events is was strongly and inversely correlated with per capita consumption of high-fructose corn syrup.
PJW
Lauretta:

This may seem like a joke but it is a very important basic statistics lesson: Correlation is not Causation. This is a fundamental flaw of so many studies that are either curve-fitting or data-mining exercises.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 10:16 am
by nedsaid
Lauretta wrote: Sun Sep 10, 2017 5:56 am I came across this paper yesterday. https://papers.ssrn.com/sol3/papers.cfm ... id=2604248
Apart from the apparent(?) spelling mistake in the title it looks like it's good stuff - one of the things that looks interesting is that compared to say momentum or moving averages rules which I was told can't really be applied to small caps because they're too volatile, this method seems to work well for small caps too.
Has anyone been looking into it? I'm not sure e.g. what the turnover would be and whether after costs this strategy is still worth considering.
I must say I have only recently begun to learn about investing - I am reading about all these methods but in practice I stick to a B&H strategy also because I think it fits my temperament better. But it's good to learn about other methods. Perhaps also to put a fraction of one's wealth in these strategies to diversify further ;-)
These type of strategies tend to work well until they don't. Sooner or later, as people wake up and start taking advantage of winning timing strategies, they stop working. The thing is, professional traders with powerful computers are working on this stuff 24-7. If something is out there to be taken advantage of, the professional traders are trying to take every bit of edge that they can.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 11:01 am
by Uncle Pennybags
nedsaid wrote: Wed Sep 13, 2017 10:16 amThese type of strategies tend to work well until they don't.
To quote Mike Tyson, "everybody has a plan until they get hit in the face".

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 11:30 am
by JBTX
nedsaid wrote: Wed Sep 13, 2017 10:16 am
Lauretta wrote: Sun Sep 10, 2017 5:56 am I came across this paper yesterday. https://papers.ssrn.com/sol3/papers.cfm ... id=2604248
Apart from the apparent(?) spelling mistake in the title it looks like it's good stuff - one of the things that looks interesting is that compared to say momentum or moving averages rules which I was told can't really be applied to small caps because they're too volatile, this method seems to work well for small caps too.
Has anyone been looking into it? I'm not sure e.g. what the turnover would be and whether after costs this strategy is still worth considering.
I must say I have only recently begun to learn about investing - I am reading about all these methods but in practice I stick to a B&H strategy also because I think it fits my temperament better. But it's good to learn about other methods. Perhaps also to put a fraction of one's wealth in these strategies to diversify further ;-)
These type of strategies tend to work well until they don't. Sooner or later, as people wake up and start taking advantage of winning timing strategies, they stop working. The thing is, professional traders with powerful computers are working on this stuff 24-7. If something is out there to be taken advantage of, the professional traders are trying to take every bit of edge that they can.
Tend to agree. I skimmed the paper and found it interesting, but have no idea how solid the data is. Don't know why they started in 1986. Even if true I would think it would be difficult for an individual to act upon. If this really has merit I would expect to see some arbitrage funds take advantage of it.

I'm not going to argue with the data, but lumber and gold aren't perfect indicators. Gold can increase during times of fear, but it also can increase due to world demand. That is primarily what happened 2001-2008 - emerging market demand drove it up. Third world citizens often don't have good access to investments, so they buy gold. In India gold is a primary investment for its citizens, and it is a big driver or world wide demand. Also, at times China's central banks buys big shares. These things seem to me to be the opposite indicators of fear.

Lumber is indeed cyclical, but can also be affected by uncontrollable factors such as pine beetles.

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 12:38 pm
by Lauretta
JBTX wrote: Wed Sep 13, 2017 11:30 am Third world citizens often don't have good access to investments, so they buy gold. In India gold is a primary investment for its citizens, and it is a big driver or world wide demand. Also, at times China's central banks buys big shares. These things seem to me to be the opposite indicators of fear.

Lumber is indeed cyclical, but can also be affected by uncontrollable factors such as pine beetles.
Interesting points, thanks for sharing :happy

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 1:15 pm
by JBTX
Lauretta wrote: Wed Sep 13, 2017 12:38 pm
JBTX wrote: Wed Sep 13, 2017 11:30 am Third world citizens often don't have good access to investments, so they buy gold. In India gold is a primary investment for its citizens, and it is a big driver or world wide demand. Also, at times China's central banks buys big shares. These things seem to me to be the opposite indicators of fear.

Lumber is indeed cyclical, but can also be affected by uncontrollable factors such as pine beetles.
Interesting points, thanks for sharing :happy
I can't find the original whitepaper from GMO on gold/demand, but here is a summary of it

https://www.valuewalk.com/2013/04/gold- ... et-demand/

Also, I found this one of the most comprehensive analyses of gold, and it happened to be a guy who was my professor about 30 years ago!

https://papers.ssrn.com/sol3/papers.cfm ... id=2078535

Re: Market timing using Gold-Lumber indicator

Posted: Wed Sep 13, 2017 1:27 pm
by Lauretta
JBTX wrote: Wed Sep 13, 2017 1:15 pm
Lauretta wrote: Wed Sep 13, 2017 12:38 pm
JBTX wrote: Wed Sep 13, 2017 11:30 am Third world citizens often don't have good access to investments, so they buy gold. In India gold is a primary investment for its citizens, and it is a big driver or world wide demand. Also, at times China's central banks buys big shares. These things seem to me to be the opposite indicators of fear.

Lumber is indeed cyclical, but can also be affected by uncontrollable factors such as pine beetles.
Interesting points, thanks for sharing :happy
I can't find the original whitepaper from GMO on gold/demand, but here is a summary of it

https://www.valuewalk.com/2013/04/gold- ... et-demand/

Also, I found this one of the most comprehensive analyses of gold, and it happened to be a guy who was my professor about 30 years ago!

https://papers.ssrn.com/sol3/papers.cfm ... id=2078535
Thanks! :happy I'll read these with interest - I know very little about gold for the moment.