My trend following strategy and experience

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 7:01 pm

FRT15 wrote:
Fri Jan 18, 2019 6:59 pm
If a change is required do you do it on the first trading day of the month? Also for the 7 months, do you include the month that just closed or is it the prior 7?
Considering that I might need to move out of stocks, I wait until the first Friday of the next month when the BLS releases the latest unemployment rate data. This limits my trading to no more than once per month.

I don't manually compute the 7 month moving average; I use Portfolio Visualizer for that. The 7 month moving average includes the most recent prior month.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Vulcan
Posts: 1025
Joined: Sat Apr 05, 2014 11:43 pm

Re: My trend following strategy and experience

Post by Vulcan » Fri Jan 18, 2019 7:14 pm

willthrill81 wrote:
Fri Jan 18, 2019 6:57 pm
Vulcan wrote:
Fri Jan 18, 2019 6:56 pm
willthrill81 wrote:
Fri Jan 18, 2019 6:28 pm
The only time a whipsaw could even occur is if the UER is above its 12 MMA, because unless that's true, I don't move out of stocks at all.
And that can't happen multiple times in a row because...
It could. And all of our stock holdings, both U.S. and international, could be confiscated and all bonds defaulted upon. If you're looking for absolute guarantees, investing isn't the place to search.
That is true, however, can't the strategy you employ conceivably lead to significant losses even in the absence of any global calamity, and in a long-term growing market?
If you torture the data long enough, it will confess to anything. ~Ronald Coase

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 7:16 pm

Vulcan wrote:
Fri Jan 18, 2019 7:14 pm
willthrill81 wrote:
Fri Jan 18, 2019 6:57 pm
Vulcan wrote:
Fri Jan 18, 2019 6:56 pm
willthrill81 wrote:
Fri Jan 18, 2019 6:28 pm
The only time a whipsaw could even occur is if the UER is above its 12 MMA, because unless that's true, I don't move out of stocks at all.
And that can't happen multiple times in a row because...
It could. And all of our stock holdings, both U.S. and international, could be confiscated and all bonds defaulted upon. If you're looking for absolute guarantees, investing isn't the place to search.
That is true, however, can't the strategy you employ conceivably lead to significant losses even in the absence of any global calamity, and in a long-term growing market?
I already said that it could.

But it would be difficult for a long-term growing market to remain under its 7 month moving average AND for the unemployment rate to remain above its 12 month moving average over the long-term.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Vulcan
Posts: 1025
Joined: Sat Apr 05, 2014 11:43 pm

Re: My trend following strategy and experience

Post by Vulcan » Fri Jan 18, 2019 7:28 pm

willthrill81 wrote:
Fri Jan 18, 2019 7:16 pm
Vulcan wrote:
Fri Jan 18, 2019 7:14 pm
willthrill81 wrote:
Fri Jan 18, 2019 6:57 pm
Vulcan wrote:
Fri Jan 18, 2019 6:56 pm
willthrill81 wrote:
Fri Jan 18, 2019 6:28 pm
The only time a whipsaw could even occur is if the UER is above its 12 MMA, because unless that's true, I don't move out of stocks at all.
And that can't happen multiple times in a row because...
It could. And all of our stock holdings, both U.S. and international, could be confiscated and all bonds defaulted upon. If you're looking for absolute guarantees, investing isn't the place to search.
That is true, however, can't the strategy you employ conceivably lead to significant losses even in the absence of any global calamity, and in a long-term growing market?
I already said that it could.

But it would be difficult for a long-term growing market to remain under its 7 month moving average AND for the unemployment rate to remain above its 12 month moving average over the long-term.
Why?
If you torture the data long enough, it will confess to anything. ~Ronald Coase

Global100
Posts: 124
Joined: Wed Dec 27, 2017 8:46 pm

Re: My trend following strategy and experience

Post by Global100 » Fri Jan 18, 2019 7:38 pm

willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Last edited by Global100 on Fri Jan 18, 2019 7:51 pm, edited 3 times in total.

gluskap
Posts: 83
Joined: Mon Nov 13, 2017 2:07 pm

Re: My trend following strategy and experience

Post by gluskap » Fri Jan 18, 2019 7:41 pm

I was also very interested in trend following but did discover there are some downsides to it. Namely whipsawing, etc. So what I do now is I am buy and hold for any existing positions. But I switch all my new contributions 100% to what the trend indicates. But since I am lazy to do all the calculations I will follow this post to help me determine when to switch my contributions lol. I am now contributing 100% new investments to bonds. Hopefully when the economy recovers then I will switch new contributions back to stocks. I feel like this is a good compromise for me.

User avatar
Vulcan
Posts: 1025
Joined: Sat Apr 05, 2014 11:43 pm

Re: My trend following strategy and experience

Post by Vulcan » Fri Jan 18, 2019 8:12 pm

gluskap wrote:
Fri Jan 18, 2019 7:41 pm
I was also very interested in trend following but did discover there are some downsides to it. Namely whipsawing, etc.
“After nearly 50 years in this business, I do not know of anybody who has done market timing successfully and consistently. I don't even know anybody who knows anybody who has done it successfully and consistently.” ~Jack Bogle
If you torture the data long enough, it will confess to anything. ~Ronald Coase

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 8:57 pm

Global100 wrote:
Fri Jan 18, 2019 7:38 pm
willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 8:59 pm

Vulcan wrote:
Fri Jan 18, 2019 7:28 pm
willthrill81 wrote:
Fri Jan 18, 2019 7:16 pm
Vulcan wrote:
Fri Jan 18, 2019 7:14 pm
willthrill81 wrote:
Fri Jan 18, 2019 6:57 pm
Vulcan wrote:
Fri Jan 18, 2019 6:56 pm

And that can't happen multiple times in a row because...
It could. And all of our stock holdings, both U.S. and international, could be confiscated and all bonds defaulted upon. If you're looking for absolute guarantees, investing isn't the place to search.
That is true, however, can't the strategy you employ conceivably lead to significant losses even in the absence of any global calamity, and in a long-term growing market?
I already said that it could.

But it would be difficult for a long-term growing market to remain under its 7 month moving average AND for the unemployment rate to remain above its 12 month moving average over the long-term.
Why?
Because if stocks are moving up, then they cannot remain under their 7 month moving average over the long-term.

I'm not sure why you're being antagonistic about this issue. I'm not trying to convince you or anyone else to follow my strategy. But many outside this thread (and in it) have expressed interest in it, which is why I posted it. I know full well that I might underperform the market and have explicitly said so, so there's no need to tell me.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Global100
Posts: 124
Joined: Wed Dec 27, 2017 8:46 pm

Re: My trend following strategy and experience

Post by Global100 » Fri Jan 18, 2019 9:23 pm

willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm
willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
Seems like if the UERs are rounded to the tenths place, then the 12mma calculation should also be rounded to the tenths place, from 3.89 to 3.9 for consistency. That rounded 3.9 then would not meet the first criterion (UER higher than the 12mma) to sell off 100% stocks - a major move. Is there a website where you get your data that publishes more precise percentages/decimals than tenths places for UERs and 12 mma?

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 9:26 pm

Global100 wrote:
Fri Jan 18, 2019 9:23 pm
willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm
willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
Seems like if the UERs are rounded to the tenths place, then the 12mma calculation should also be rounded to the tenths place, from 3.89 to 3.9 for consistency. That rounded 3.9 then would not meet the first criterion (UER higher than the 12mma) to sell off 100% stocks. Is there a website where you get your data that publishes more precise numbers/decimals for UERs and 12 mma?
I checked again, and it's not a rounding error. As calculated by the Bureau of Labor Statistics at this page, the 12 MMA for the UER is 3.8917 (this includes data from January through December of 2018). The December UER was 3.9, which is above the 12 MMA.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Rowan Oak
Posts: 383
Joined: Mon May 09, 2016 2:11 pm
Location: Yoknapatawpha

Re: My trend following strategy and experience

Post by Rowan Oak » Fri Jan 18, 2019 9:28 pm

Thank you for posting about your process. Will be following along.
“If you can get good at destroying your own wrong ideas, that is a great gift.” – Charlie Munger

User avatar
One Ping
Posts: 773
Joined: Thu Sep 24, 2015 4:53 pm

Re: My trend following strategy and experience

Post by One Ping » Fri Jan 18, 2019 9:43 pm

willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm
willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
The BLS data site I found only shows data to 1 significant figure (https://data.bls.gov/timeseries/lns14000000). Where do you find UER data to 2 significant figures where rounding would matter?
"Re-verify our range to target ... one ping only."

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 9:45 pm

One Ping wrote:
Fri Jan 18, 2019 9:43 pm
willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm
willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
The BLS data site I found only shows data to 1 significant figure (https://data.bls.gov/timeseries/lns14000000). Where do you find UER data to 2 significant figures where rounding would matter?
I didn't pay close enough attention to Global100's comment. I mistakenly thought it was due to a rounding error, but it wasn't. The Dec. UER is .01% above the 12 MMA for UER, turning 'on' that signal.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
One Ping
Posts: 773
Joined: Thu Sep 24, 2015 4:53 pm

Re: My trend following strategy and experience

Post by One Ping » Fri Jan 18, 2019 9:50 pm

willthrill81 wrote:
Fri Jan 18, 2019 9:45 pm
One Ping wrote:
Fri Jan 18, 2019 9:43 pm
willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm
willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
The BLS data site I found only shows data to 1 significant figure (https://data.bls.gov/timeseries/lns14000000). Where do you find UER data to 2 significant figures where rounding would matter?
I didn't pay close enough attention to Global100's comment. I mistakenly thought it was due to a rounding error, but it wasn't. The Dec. UER is .01% above the 12 MMA for UER, turning 'on' that signal.
got it. thx. :beer
"Re-verify our range to target ... one ping only."

User avatar
Vulcan
Posts: 1025
Joined: Sat Apr 05, 2014 11:43 pm

Re: My trend following strategy and experience

Post by Vulcan » Fri Jan 18, 2019 9:55 pm

willthrill81 wrote:
Fri Jan 18, 2019 8:59 pm
Vulcan wrote:
Fri Jan 18, 2019 7:28 pm
willthrill81 wrote:
Fri Jan 18, 2019 7:16 pm
But it would be difficult for a long-term growing market to remain under its 7 month moving average AND for the unemployment rate to remain above its 12 month moving average over the long-term.
Why?
Because if stocks are moving up, then they cannot remain under their 7 month moving average over the long-term.

I'm not sure why you're being antagonistic about this issue. I'm not trying to convince you or anyone else to follow my strategy.
What if they are moving up in short bursts? Slowly down 20 month in a row, then sharply up. Repeat.

Just trying to get to the bottom of the argument - is there a fundamental case underlying trend following separate from the backtesting heuristics?

If not, how is this different from any number of other timing strategies that exist - or can be designed to pass the backtest? You are basically saying that you want to minimize the drawdown risk, and you have a mechanical device that allows you to do that. Yet I see no physical explanation for why it should work in the future.
If you torture the data long enough, it will confess to anything. ~Ronald Coase

User avatar
raven15
Posts: 382
Joined: Sun Nov 30, 2014 8:01 pm

Re: My trend following strategy and experience

Post by raven15 » Fri Jan 18, 2019 9:59 pm

@willthrill
I have two snarky comments.

First, I remember when you first started posting here. First you were all about gold and how amazing it was. Then you thought small cap value was the hot asset, but just after that you determined mid cap value was the bees knees. Then I stopped reading for a few months, and came back and saw you were a trend follower and actually lol'ed. You probably thought it was news. But I could tell you were a trend follower the whole time!

Second, I have recently been wondering about the anomalous out performance of trend following in general and how it can persist. If it continues forever trend followers will become super wealthy at the expense of all the other people, even buy and holders. How can that work? So I was wondering what it would take to shake trend followers free. I imagine it would be a decade or two similar to the past year, with wild swings across all time scales. That one didn't come out snarky, and I am a little disappointed.

Couple additions: you were also strongly against international investing of any kind, another data point pointing to your trend following nature. Second, if you regularly end up with a different asset class in every account that could be a great way to reduce the risk of the strategy.
Last edited by raven15 on Fri Jan 18, 2019 10:20 pm, edited 1 time in total.
It's Time. Adding Interest.

User avatar
Vulcan
Posts: 1025
Joined: Sat Apr 05, 2014 11:43 pm

Re: My trend following strategy and experience

Post by Vulcan » Fri Jan 18, 2019 10:03 pm

raven15 wrote:
Fri Jan 18, 2019 9:59 pm
@willthrill
I have two snarky comments.

First, I remember when you first started posting here. First you were all about gold and how amazing it was. Then you thought small cap value was the hot asset, but just after that you determined mid cap value was the bees knees. Then I stopped reading for a few months, and came back and saw you were a trend follower and actually lol'ed. You probably thought it was news. But I could tell you were a trend follower the whole time!

Second, I have recently been wondering about the anomalous out performance of trend following in general and how it can persist. If it continues forever trend followers will become super wealthy at the expense of all the other people, even buy and holders. How can that work? So I was wondering what it would take to shake trend followers free. I imagine it would be a decade or two similar to the past year, with wild swings across all time scales. That one didn't come out snarky, and I am a little disappointed.
:twisted:
If you torture the data long enough, it will confess to anything. ~Ronald Coase

Global100
Posts: 124
Joined: Wed Dec 27, 2017 8:46 pm

Re: My trend following strategy and experience

Post by Global100 » Fri Jan 18, 2019 10:04 pm

willthrill81 wrote:
Fri Jan 18, 2019 9:26 pm
Global100 wrote:
Fri Jan 18, 2019 9:23 pm
willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm
willthrill81 wrote:
Thu Jan 17, 2019 7:16 pm
The UER is .01% higher than it's 12 month moving average
Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
Seems like if the UERs are rounded to the tenths place, then the 12mma calculation should also be rounded to the tenths place, from 3.89 to 3.9 for consistency. That rounded 3.9 then would not meet the first criterion (UER higher than the 12mma) to sell off 100% stocks. Is there a website where you get your data that publishes more precise numbers/decimals for UERs and 12 mma?
I checked again, and it's not a rounding error. As calculated by the Bureau of Labor Statistics at this page, the 12 MMA for the UER is 3.8917 (this includes data from January through December of 2018). The December UER was 3.9, which is above the 12 MMA.
Not finding 3.8917 on their data web page. But if I add up all the UERs (which were rounded to the tenths place) equals 46.7; divide by 12 months is 3.891666666 (3.8917). If UERs were rounded to tenths and used in the calculations for the average, then the resulting answer is most likely not precise in the hundredths place.

Maybe this strategy was specifically developed by always rounding to the tenths place for the UERs and purposely including the hundredths place for 12mma. If so, I can accept that. Really would like to get this calculation part of the strategy understood, as it seems very critical to whether one could be selling or buying 100% stocks that month. Thanks.

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 10:10 pm

Global100 wrote:
Fri Jan 18, 2019 10:04 pm
willthrill81 wrote:
Fri Jan 18, 2019 9:26 pm
Global100 wrote:
Fri Jan 18, 2019 9:23 pm
willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm

Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
Seems like if the UERs are rounded to the tenths place, then the 12mma calculation should also be rounded to the tenths place, from 3.89 to 3.9 for consistency. That rounded 3.9 then would not meet the first criterion (UER higher than the 12mma) to sell off 100% stocks. Is there a website where you get your data that publishes more precise numbers/decimals for UERs and 12 mma?
I checked again, and it's not a rounding error. As calculated by the Bureau of Labor Statistics at this page, the 12 MMA for the UER is 3.8917 (this includes data from January through December of 2018). The December UER was 3.9, which is above the 12 MMA.
Not finding 3.8917 on their data web page. But if I add up all the UERs (which were rounded to the tenths place) equals 46.7; divide by 12 months is 3.891666666 (3.8917). If UERs were rounded to tenths and used in the calculations for the average, then the resulting answer is most likely not precise in the hundredths place.

Maybe this strategy was specifically developed by always rounding to the tenths place for the UERS and purposely including the hundredths place for 12mma. If so, I can accept that. Really would like to get this calculation part of the strategy understood, as it seems very critical to whether one could be selling or buying 100% stocks that month. Thanks.
The point is that 3.90 is higher than 3.89.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Vulcan
Posts: 1025
Joined: Sat Apr 05, 2014 11:43 pm

Re: My trend following strategy and experience

Post by Vulcan » Fri Jan 18, 2019 10:12 pm

Global100 wrote:
Fri Jan 18, 2019 10:04 pm
Maybe this strategy was specifically developed by always rounding to the tenths place for the UERs and purposely including the hundredths place for 12mma. If so, I can accept that. Really would like to get this calculation part of the strategy understood, as it seems very critical to whether one could be selling or buying 100% stocks that month. Thanks.
Don't you see how ridiculous all of this sounds right there?
If you torture the data long enough, it will confess to anything. ~Ronald Coase

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 10:14 pm

raven15 wrote:
Fri Jan 18, 2019 9:59 pm
@willthrill
I have two snarky comments.

First, I remember when you first started posting here. First you were all about gold and how amazing it was. Then you thought small cap value was the hot asset, but just after that you determined mid cap value was the bees knees. Then I stopped reading for a few months, and came back and saw you were a trend follower and actually lol'ed. You probably thought it was news. But I could tell you were a trend follower the whole time!

Second, I have recently been wondering about the anomalous out performance of trend following in general and how it can persist. If it continues forever trend followers will become super wealthy at the expense of all the other people, even buy and holders. How can that work? So I was wondering what it would take to shake trend followers free. I imagine it would be a decade or two similar to the past year, with wild swings across all time scales. That one didn't come out snarky, and I am a little disappointed.
To be honest, when I first came to this forum, I was still getting me 'sea legs', as it were, despite knowing a fair amount about personal finance as a distinct topic. I've never been a gold bug, although I do think that it's ability to help stabilize portfolios is too quickly dismissed by many. And yes, if I were a buy-and-holder, I'd definitely tilt toward SCV.

But I know more now than I did then. Maybe it will help me, and maybe it won't. We'll see. I've tried to be upfront and honest about the whole process, and I won't apologize for that.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
raven15
Posts: 382
Joined: Sun Nov 30, 2014 8:01 pm

Re: My trend following strategy and experience

Post by raven15 » Fri Jan 18, 2019 10:25 pm

willthrill81 wrote:
Fri Jan 18, 2019 10:14 pm
raven15 wrote:
Fri Jan 18, 2019 9:59 pm
@willthrill
I have two snarky comments.

First, I remember when you first started posting here. First you were all about gold and how amazing it was. Then you thought small cap value was the hot asset, but just after that you determined mid cap value was the bees knees. Then I stopped reading for a few months, and came back and saw you were a trend follower and actually lol'ed. You probably thought it was news. But I could tell you were a trend follower the whole time!

Second, I have recently been wondering about the anomalous out performance of trend following in general and how it can persist. If it continues forever trend followers will become super wealthy at the expense of all the other people, even buy and holders. How can that work? So I was wondering what it would take to shake trend followers free. I imagine it would be a decade or two similar to the past year, with wild swings across all time scales. That one didn't come out snarky, and I am a little disappointed.
To be honest, when I first came to this forum, I was still getting me 'sea legs', as it were, despite knowing a fair amount about personal finance as a distinct topic. I've never been a gold bug, although I do think that it's ability to help stabilize portfolios is too quickly dismissed by many. And yes, if I were a buy-and-holder, I'd definitely tilt toward SCV.

But I know more now than I did then. Maybe it will help me, and maybe it won't. We'll see. I've tried to be upfront and honest about the whole process, and I won't apologize for that.
FYI I was not insulting or injuring you. I post snarky comments for their entertainment value because usually all the good snarkfree ones are already taken by smarter and faster people.
It's Time. Adding Interest.

furnace
Posts: 330
Joined: Tue Oct 20, 2015 3:38 pm

Re: My trend following strategy and experience

Post by furnace » Fri Jan 18, 2019 10:30 pm

OP, so you want to be a quant? I hope you're doing this with "play" money. Get it right before you go big. :mrgreen:

cheezit
Posts: 260
Joined: Sat Jul 14, 2018 7:28 pm

Re: My trend following strategy and experience

Post by cheezit » Fri Jan 18, 2019 10:36 pm

The only pathological input to this algorithm that I can think of off the top of my head would be unemployment and the stock market both sort of oscillating around their moving averages but out-of-phase, causing repeated whipsaw losses (proportional to the magnitude of the market oscillation) in an overall sideways market. Not sure how likely that is to ever happen, or whether it's even worth thinking about.

Another curiosity comes to mind: the December unemployment numbers went up if you were looking at U3, but down if you were looking at some other measures (eg. prime-age male labor force participation, and I think several other measures). The algorithm might be weak against an economic recovery where jobs are being created and more people are legitimately working but the headline unemployment number (usually U3) is going up due to discouraged workers returning to the labor force.

bikechuck
Posts: 564
Joined: Sun Aug 16, 2015 9:22 pm

Re: My trend following strategy and experience

Post by bikechuck » Fri Jan 18, 2019 10:40 pm

Hi Willthrill, thank you for starting this thread and best wishes for successful results going forward.

I noticed your positive comments about TIAA Real Estate and that you have added to your TREA holdings. I own this in my simple buy and hold portfolio and it was a good performer in 2018 but at times I consider selling it in the interest of simplicity.

Can you share your reasons for optimism for TIAA Real Estate at this time?

Thanks so much for any thoughts about this that you care to share.

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 11:24 pm

cheezit wrote:
Fri Jan 18, 2019 10:36 pm
The only pathological input to this algorithm that I can think of off the top of my head would be unemployment and the stock market both sort of oscillating around their moving averages but out-of-phase, causing repeated whipsaw losses (proportional to the magnitude of the market oscillation) in an overall sideways market. Not sure how likely that is to ever happen, or whether it's even worth thinking about.
I don't believe that the unemployment rate hasn't tended to oscillate very much. It's generally been trending one way or the other.
cheezit wrote:
Fri Jan 18, 2019 10:36 pm
Another curiosity comes to mind: the December unemployment numbers went up if you were looking at U3, but down if you were looking at some other measures (eg. prime-age male labor force participation, and I think several other measures). The algorithm might be weak against an economic recovery where jobs are being created and more people are legitimately working but the headline unemployment number (usually U3) is going up due to discouraged workers returning to the labor force.
To be sure, the UER is not a perfect measure. But historically, it's been an excellent recession indicator, as those in the Fed have publicly stated. Of course, I'm only using it as a sort of filter to minimize the whipsaws that would otherwise occur if I simply followed the moving average of the stock indexes.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 11:28 pm

bikechuck wrote:
Fri Jan 18, 2019 10:40 pm
Hi Willthrill, thank you for starting this thread and best wishes for successful results going forward.

I noticed your positive comments about TIAA Real Estate and that you have added to your TREA holdings. I own this in my simple buy and hold portfolio and it was a good performer in 2018 but at times I consider selling it in the interest of simplicity.

Can you share your reasons for optimism for TIAA Real Estate at this time?

Thanks so much for any thoughts about this that you care to share.
I won't get into the specifics for the sake of keeping the thread on track. Broadly speaking, TREA directly holds real estate, and it's growth has consequently been far steadier and consistent than REITs. And due to the operation of TREA, it's so easy to time it that TIAA has limited how much exposure you can have in the fund and your number of buys to just one per quarter, IIRC. You can read more about them in various other threads, such as this one.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
ResearchMed
Posts: 9372
Joined: Fri Dec 26, 2008 11:25 pm

Re: My trend following strategy and experience

Post by ResearchMed » Fri Jan 18, 2019 11:32 pm

willthrill81 wrote:
Fri Jan 18, 2019 11:28 pm
bikechuck wrote:
Fri Jan 18, 2019 10:40 pm
Hi Willthrill, thank you for starting this thread and best wishes for successful results going forward.

I noticed your positive comments about TIAA Real Estate and that you have added to your TREA holdings. I own this in my simple buy and hold portfolio and it was a good performer in 2018 but at times I consider selling it in the interest of simplicity.

Can you share your reasons for optimism for TIAA Real Estate at this time?

Thanks so much for any thoughts about this that you care to share.
I won't get into the specifics for the sake of keeping the thread on track. Broadly speaking, TREA directly holds real estate, and it's growth has consequently been far steadier and consistent than REITs. And due to the operation of TREA, it's so easy to time it that TIAA has limited how much exposure you can have in the fund and your number of buys to just one per quarter, IIRC. You can read more about them in various other threads, such as this one.
It is "selling" that is restricted to once per quarter.

Direct purchases are restricted once one has reached a ceiling, but most can still get more in with some sorts of transfers.

RM
This signature is a placebo. You are in the control group.

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Fri Jan 18, 2019 11:35 pm

ResearchMed wrote:
Fri Jan 18, 2019 11:32 pm
willthrill81 wrote:
Fri Jan 18, 2019 11:28 pm
bikechuck wrote:
Fri Jan 18, 2019 10:40 pm
Hi Willthrill, thank you for starting this thread and best wishes for successful results going forward.

I noticed your positive comments about TIAA Real Estate and that you have added to your TREA holdings. I own this in my simple buy and hold portfolio and it was a good performer in 2018 but at times I consider selling it in the interest of simplicity.

Can you share your reasons for optimism for TIAA Real Estate at this time?

Thanks so much for any thoughts about this that you care to share.
I won't get into the specifics for the sake of keeping the thread on track. Broadly speaking, TREA directly holds real estate, and it's growth has consequently been far steadier and consistent than REITs. And due to the operation of TREA, it's so easy to time it that TIAA has limited how much exposure you can have in the fund and your number of buys to just one per quarter, IIRC. You can read more about them in various other threads, such as this one.
It is "selling" that is restricted to once per quarter.

Direct purchases are restricted once one has reached a ceiling, but most can still get more in with some sorts of transfers.

RM
I figured I would get it wrong. Thanks for the correction. :beer
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

pascalwager
Posts: 1568
Joined: Mon Oct 31, 2011 8:36 pm

Re: My trend following strategy and experience

Post by pascalwager » Sat Jan 19, 2019 2:59 am

Global100 wrote:
Fri Jan 18, 2019 10:04 pm
willthrill81 wrote:
Fri Jan 18, 2019 9:26 pm
Global100 wrote:
Fri Jan 18, 2019 9:23 pm
willthrill81 wrote:
Fri Jan 18, 2019 8:57 pm
Global100 wrote:
Fri Jan 18, 2019 7:38 pm

Is this how the 12 month moving average was calculated?

2018 unemployment data
Jan 4.1
Feb 4.1
Mar 4.0
Apr 3.9
May 3.8
June 4.0
July 3.9
Aug 3.8
Sep 3.7
Oct 3.8
Nov 3.7
Dec 3.9

Average rounds to 3.89

If some of the numbers used above were rounded down (e.g. 4.14 to 4.1) and December's 3.9 had been rounded up, then it's possible the latest UER is lower than the twelve month moving average.
Yes, it's due to a rounding issue.
Seems like if the UERs are rounded to the tenths place, then the 12mma calculation should also be rounded to the tenths place, from 3.89 to 3.9 for consistency. That rounded 3.9 then would not meet the first criterion (UER higher than the 12mma) to sell off 100% stocks. Is there a website where you get your data that publishes more precise numbers/decimals for UERs and 12 mma?
I checked again, and it's not a rounding error. As calculated by the Bureau of Labor Statistics at this page, the 12 MMA for the UER is 3.8917 (this includes data from January through December of 2018). The December UER was 3.9, which is above the 12 MMA.
Not finding 3.8917 on their data web page. But if I add up all the UERs (which were rounded to the tenths place) equals 46.7; divide by 12 months is 3.891666666 (3.8917). If UERs were rounded to tenths and used in the calculations for the average, then the resulting answer is most likely not precise in the hundredths place.

Maybe this strategy was specifically developed by always rounding to the tenths place for the UERs and purposely including the hundredths place for 12mma. If so, I can accept that. Really would like to get this calculation part of the strategy understood, as it seems very critical to whether one could be selling or buying 100% stocks that month. Thanks.
I think you do understand the "calculation part of the strategy". When you round off the 12-month average the result is 3.9--same as the December value (3.9). So I don't see a recession signal here. But I guess Will thought the stock indices average was a strong enough signal to make a move to fixed income.

hilink73
Posts: 332
Joined: Tue Sep 20, 2016 3:29 pm

Re: My trend following strategy and experience

Post by hilink73 » Sat Jan 19, 2019 3:03 am

willthrill81 wrote:
Fri Jan 18, 2019 2:12 pm
aristotelian wrote:
Fri Jan 18, 2019 12:59 pm
I don't get the logic of trend following. If you were to time the market, wouldn't you want to get out at the peak, when the market is still trending up, instead of when the market has already started going down? Personally, if I were to time the market, I would be more inclined to look at valuation rather than trend, and try to sell before the drop rather than in the middle of it.
I know of no one with any measure of credibility who even claims to be able to call the top before there is any downward movement. Trend following has worked over the long-term by largely preventing the worst of the market falls. You partially 'pay' for this with whipsaws, where you sell at a lower price than you buy back in at but are 'vindicated' when the opposite happens in a substantial way. That's part of the reason why many don't stick with trend following: it may take years for the 'vindication' to occur.

Valuations have been shown to be terrible market timing signals because they are not reliably mean reverting.
aristotelian wrote:
Fri Jan 18, 2019 12:59 pm
Did you get out of stocks on Jan 1?
No. The BLS data weren't released until later, and as I noted above, I didn't notice the change in the UER until earlier this week, so I was a little late to make the switch. I have reminders set up now to prevent this going forward.
https://www.bls.gov/schedule/news_release/empsit.htm

Code: Select all

Reference Month 	Release Date 	Release Time
September 2018		Oct. 05, 2018 	08:30 AM
October 2018		Nov. 02, 2018 	08:30 AM
November 2018		Dec. 07, 2018 	08:30 AM
December 2018		Jan. 04, 2019 	08:30 AM
January 2019		Feb. 01, 2019 	08:30 AM
February 2019		Mar. 08, 2019 	08:30 AM
March 2019		Apr. 05, 2019 	08:30 AM
April 2019		May 03, 2019 	08:30 AM
May 2019		Jun. 07, 2019 	08:30 AM
June 2019		Jul. 05, 2019 	08:30 AM
July 2019		Aug. 02, 2019 	08:30 AM
August 2019		Sep. 06, 2019 	08:30 AM
September 2019		Oct. 04, 2019 	08:30 AM
October 2019		Nov. 01, 2019 	08:30 AM
November 2019		Dec. 06, 2019 	08:30 AM
I have put those dates in my calendar...

AlohaJoe
Posts: 4943
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: My trend following strategy and experience

Post by AlohaJoe » Sat Jan 19, 2019 3:29 am

You've got over 1,000 replies to go before you catch up to the greatest trend following post/flamewar in Bogleheads's history

viewtopic.php?t=27460

Good luck on your adventure ;)

long_gamma
Posts: 352
Joined: Mon Sep 14, 2015 4:13 pm

Re: My trend following strategy and experience

Post by long_gamma » Sat Jan 19, 2019 6:41 am

willthrill81 wrote:
Fri Jan 18, 2019 10:10 pm

The point is that 3.90 is higher than 3.89.
Underlying data is quoted in tenth's (3.9, 3.8, 3.7 ), not in 100'ths (3.89, 3.91, 3.78 etc). Your indicator calculation can not be more precise than the underlying source. In current scenario both underlying and indicator has the same value of 3.9. Don't know what is your system rules when they have the same value.

Also, it is not "trend-following" as followed by most CTA's. No one will have such hard cut-offs. Their position is dictated by the signal strength. Here in this case signal deteriorated marginally, so they reduce their position marginally.
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

long_gamma
Posts: 352
Joined: Mon Sep 14, 2015 4:13 pm

Re: My trend following strategy and experience

Post by long_gamma » Sat Jan 19, 2019 7:19 am

AlohaJoe wrote:
Sat Jan 19, 2019 3:29 am
You've got over 1,000 replies to go before you catch up to the greatest trend following post/flamewar in Bogleheads's history

viewtopic.php?t=27460

Good luck on your adventure ;)
There was one old thread from M* Vanguard Diehard forum (prior version of Bogleheads) in early 2000's or late 90's which was full of long entertaining and snarky comments. I believe it had highest number of posts in the Diehard forum.
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

foo.c
Posts: 62
Joined: Sat Jul 15, 2017 4:55 pm

Re: My trend following strategy and experience

Post by foo.c » Sat Jan 19, 2019 8:25 am

One thing that occurs to me is that how much can you trust the UE numbers in real time? Aren't they revised frequently?

I know the back tests show a great indicator in back testing, but how about in real time?

long_gamma
Posts: 352
Joined: Mon Sep 14, 2015 4:13 pm

Re: My trend following strategy and experience

Post by long_gamma » Sat Jan 19, 2019 9:58 am

foo.c wrote:
Sat Jan 19, 2019 8:25 am
One thing that occurs to me is that how much can you trust the UE numbers in real time? Aren't they revised frequently?

I know the back tests show a great indicator in back testing, but how about in real time?
http://www.philosophicaleconomics.com/2016/02/uetrend/

Philosophical economics tests this both on revised and unrevised numbers. Unrevised numbers actually had slightly better results.
"Everyone has a plan 'till they get punched in the mouth." --Mike Tyson

KlangFool
Posts: 14177
Joined: Sat Oct 11, 2008 12:35 pm

Re: My trend following strategy and experience

Post by KlangFool » Sat Jan 19, 2019 10:13 am

OP,

A) Are you doing this for 100% of your portfolio or just part of your portfolio?

B) Are they in the tax-advantaged account? If not, how do you account for the tax inefficiency? Aka, the amount of taxes that you have to pay?

KlangFool

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Sat Jan 19, 2019 10:33 am

long_gamma wrote:
Sat Jan 19, 2019 6:41 am
willthrill81 wrote:
Fri Jan 18, 2019 10:10 pm

The point is that 3.90 is higher than 3.89.
Underlying data is quoted in tenth's (3.9, 3.8, 3.7 ), not in 100'ths (3.89, 3.91, 3.78 etc). Your indicator calculation can not be more precise than the underlying source. In current scenario both underlying and indicator has the same value of 3.9. Don't know what is your system rules when they have the same value.
My system calls me for me to do what precisely what I've done.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Sat Jan 19, 2019 10:37 am

KlangFool wrote:
Sat Jan 19, 2019 10:13 am
OP,

A) Are you doing this for 100% of your portfolio or just part of your portfolio?
100% of our portfolio. Our savings rate is high enough (just over 50%) that we don't need strong returns to reach financial independence in a reasonable amount of time, so somewhat poorer performance than a balanced fund would not be a real problem for us. But I'm confident that over the long-term, the portfolio will do at least as well as an 85/15 AA.
KlangFool wrote:
Sat Jan 19, 2019 10:13 am
B) Are they in the tax-advantaged account? If not, how do you account for the tax inefficiency? Aka, the amount of taxes that you have to pay?
100% in tax-advantaged accounts.

That being said, you can use futures contracts to zero your stock position in a taxable account rather than sell shares and incur capital gains.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

KlangFool
Posts: 14177
Joined: Sat Oct 11, 2008 12:35 pm

Re: My trend following strategy and experience

Post by KlangFool » Sat Jan 19, 2019 10:45 am

willthrill81 wrote:
Sat Jan 19, 2019 10:37 am
KlangFool wrote:
Sat Jan 19, 2019 10:13 am
OP,

A) Are you doing this for 100% of your portfolio or just part of your portfolio?
100% of our portfolio. Our savings rate is high enough (just over 50%) that we don't need strong returns to reach financial independence in a reasonable amount of time, so somewhat poorer performance than a balanced fund would not be a real problem for us. But I'm confident that over the long-term, the portfolio will do at least as well as an 85/15 AA.
KlangFool wrote:
Sat Jan 19, 2019 10:13 am
B) Are they in the tax-advantaged account? If not, how do you account for the tax inefficiency? Aka, the amount of taxes that you have to pay?
100% in tax-advantaged accounts.

That being said, you can use futures contracts to zero your stock position in a taxable account rather than sell shares and incur capital gains.
willthrill81,

1) Good luck!

2) I keep 40% of my portfolio at Wellington fund. I will let those folks do this kind of stuff. They had survived for 80+ years. I trust that they have a system that works.

KlangFool

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Sat Jan 19, 2019 10:46 am

KlangFool wrote:
Sat Jan 19, 2019 10:45 am
willthrill81 wrote:
Sat Jan 19, 2019 10:37 am
KlangFool wrote:
Sat Jan 19, 2019 10:13 am
OP,

A) Are you doing this for 100% of your portfolio or just part of your portfolio?
100% of our portfolio. Our savings rate is high enough (just over 50%) that we don't need strong returns to reach financial independence in a reasonable amount of time, so somewhat poorer performance than a balanced fund would not be a real problem for us. But I'm confident that over the long-term, the portfolio will do at least as well as an 85/15 AA.
KlangFool wrote:
Sat Jan 19, 2019 10:13 am
B) Are they in the tax-advantaged account? If not, how do you account for the tax inefficiency? Aka, the amount of taxes that you have to pay?
100% in tax-advantaged accounts.

That being said, you can use futures contracts to zero your stock position in a taxable account rather than sell shares and incur capital gains.
willthrill81,

1) Good luck!

2) I keep 40% of my portfolio at Wellington fund. I will let those folks do this kind of stuff. They had survived for 80+ years. I trust that they have a system that works.

KlangFool
Thanks!

I think that Wellington is a fine fund and perfectly acceptable all-in-one choice.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
BlueEars
Posts: 3788
Joined: Sat Mar 10, 2007 12:15 am
Location: West Coast

Re: My trend following strategy and experience

Post by BlueEars » Sat Jan 19, 2019 10:51 am

Willthrill, my congratulations on having the guts to share in a forum which is pretty antagonistic (on average) to trend following. I've been using my own home brew system of trend following since 2009 after going through that bad decline and not selling. This system has so far not called for a trade out of equities. It is designed to only trade out in extremis such as late 1929 which I have backtested too.

After reading that Philosophical Investor article I added the unemployment rate as one component to my spreadsheet. It is a small factor (equivalent to a very high PE10 factor) with the largest factor in my approach being the yield curve. Anyway, I look for 3 continuous months of unemployment above the 12 month moving average before calling this factor a negative. My backtest with modern Fed data goes back to the 1950's. Just thought I'd mention this for your future consideration.

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Sat Jan 19, 2019 11:00 am

BlueEars wrote:
Sat Jan 19, 2019 10:51 am
Willthrill, my congratulations on having the guts to share in a forum which is pretty antagonistic (on average) to trend following. I've been using my own home brew system of trend following since 2009 after going through that bad decline and not selling. This system has so far not called for a trade out of equities. It is designed to only trade out in extremis such as late 1929 which I have backtested too.

After reading that Philosophical Investor article I added the unemployment rate as one component to my spreadsheet. It is a small factor (equivalent to a very high PE10 factor) with the largest factor in my approach being the yield curve. Anyway, I look for 3 continuous months of unemployment above the 12 month moving average before calling this factor a negative. My backtest with modern Fed data goes back to the 1950's. Just thought I'd mention this for your future consideration.
Thanks.

I'm surprised that using 3 months of above the 12 MMA as a filter signal (I'm guessing that's how you used it) would have worked well, considering that the lead time between the UER crossing its 12 MMA and the onset of the recession has historically averaged 3.5 months.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
BlueEars
Posts: 3788
Joined: Sat Mar 10, 2007 12:15 am
Location: West Coast

Re: My trend following strategy and experience

Post by BlueEars » Sat Jan 19, 2019 11:10 am

Keep in mind this is one factor in a multifactor approach. For instance, yield curve inversion may cause a switch before the unemployment rate does. Also I do not quite use a moving average approach on the SP500 but something rather different. The numbers work out and that is all I should care about. There may be a bit more loss before switching out but the overall CAGR works better from 1950-now.

User avatar
Horton
Posts: 390
Joined: Mon Jan 21, 2008 3:53 pm

Re: My trend following strategy and experience

Post by Horton » Sat Jan 19, 2019 12:31 pm

For those stumbling across this strategy for the first time, I have provided some words of caution below.

1) Philosophical Economics points out that backtests of many/most (all?) trend following strategies assume no bid-ask spreads, or other commissions. Why? Because this isn't much of an issue any longer. However, "the average spread from 1928 to 1999 was north of .60%, a number that would have completely destroyed any strategy that traded daily–and that would have significantly impaired strategies that traded monthly or even quarterly." Factor this into the backtests and the results are much less optimistic.

2) Rekenthaler’s Rule: "If the bozos know about it, it doesn’t work anymore."

3) As ERN notes, "the momentum signal also creates a lot of false signals! For short-term market fluctuations, this signal sucks!" So, you need to be committed to this strategy for the long-term.

4) "Timing luck" looms large in many trend following strategies. For example, executing trades on the last day of the month versus the first day of the month may lead to significantly different outcomes. Executing trades intra-day (i.e., using ETFs) versus at the end of the day (i.e., using mutual funds) can also lead to different outcomes. Not executing trades because you are on vacation, ill, or timid can also lead to different outcomes. It would be ideal to automate this, but that's likely not possible without additional fees that would erode much (if any) of the value of the strategy.

All this said, I acknowledge that there is plenty of evidence that momentum / trend following strategies have worked in the past and there are viable behavioral theories why it may persist in the future. Each investor needs to develop a strategy and stick with it. For many of the reasons above, I have concluded that this strategy is not right for me at this time.

@willthrill81 - please do not consider any of the above comments a critique of your personal choice. It sounds like you have thought through this in detail and are prepared to stick with it. I provide the comments above for the consideration of those who stumble across this topic for the first time. The basic message here is don't try it unless you have done your research, established a system, and are prepared to stick with it.
🏃 since 2005

User avatar
hdas
Posts: 1272
Joined: Thu Jun 11, 2015 8:24 am

Re: My trend following strategy and experience

Post by hdas » Sat Jan 19, 2019 12:44 pm

Horton wrote:
Sat Jan 19, 2019 12:31 pm
For those stumbling across this strategy for the first time, I have provided some words of caution below.

1) Philosophical Economics points out that backtests of many/most (all?) trend following strategies assume no bid-ask spreads, or other commissions. Why? Because this isn't much of an issue any longer. However, "the average spread from 1928 to 1999 was north of .60%, a number that would have completely destroyed any strategy that traded daily–and that would have significantly impaired strategies that traded monthly or even quarterly." Factor this into the backtests and the results are much less optimistic.

2) Rekenthaler’s Rule: "If the bozos know about it, it doesn’t work anymore."

3) As ERN notes, "the momentum signal also creates a lot of false signals! For short-term market fluctuations, this signal sucks!" So, you need to be committed to this strategy for the long-term.

4) "Timing luck" looms large in many trend following strategies. For example, executing trades on the last day of the month versus the first day of the month may lead to significantly different outcomes. Executing trades intra-day (i.e., using ETFs) versus at the end of the day (i.e., using mutual funds) can also lead to different outcomes. Not executing trades because you are on vacation, ill, or timid can also lead to different outcomes. It would be ideal to automate this, but that's likely not possible without additional fees that would erode much (if any) of the value of the strategy.
Well said, specially when you also add careless implementation and subjective timing, which seems to be part of the OP process given that he stated that:
willthrill81 wrote:
Thu Jan 17, 2019 11:37 pm
I was actually a bit late this month making by shift because I just glanced at the UER instead of actually checking out the moving average, when I discovered that the UER crossed its 12 MMA by .01%. I've got reminders set up now to check this the day the BLS releases the data. Henceforth, I'll make whatever trades I need to on that day of the month.
This type of behavior changes the whole distribution and also shows that the process is not objective at all.

Cheers :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

User avatar
Topic Author
willthrill81
Posts: 14007
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: My trend following strategy and experience

Post by willthrill81 » Sat Jan 19, 2019 1:09 pm

hdas wrote:
Sat Jan 19, 2019 12:44 pm
Horton wrote:
Sat Jan 19, 2019 12:31 pm
For those stumbling across this strategy for the first time, I have provided some words of caution below.

1) Philosophical Economics points out that backtests of many/most (all?) trend following strategies assume no bid-ask spreads, or other commissions. Why? Because this isn't much of an issue any longer. However, "the average spread from 1928 to 1999 was north of .60%, a number that would have completely destroyed any strategy that traded daily–and that would have significantly impaired strategies that traded monthly or even quarterly." Factor this into the backtests and the results are much less optimistic.

2) Rekenthaler’s Rule: "If the bozos know about it, it doesn’t work anymore."

3) As ERN notes, "the momentum signal also creates a lot of false signals! For short-term market fluctuations, this signal sucks!" So, you need to be committed to this strategy for the long-term.

4) "Timing luck" looms large in many trend following strategies. For example, executing trades on the last day of the month versus the first day of the month may lead to significantly different outcomes. Executing trades intra-day (i.e., using ETFs) versus at the end of the day (i.e., using mutual funds) can also lead to different outcomes. Not executing trades because you are on vacation, ill, or timid can also lead to different outcomes. It would be ideal to automate this, but that's likely not possible without additional fees that would erode much (if any) of the value of the strategy.
Well said, specially when you also add careless implementation and subjective timing, which seems to be part of the OP process given that he stated that:
willthrill81 wrote:
Thu Jan 17, 2019 11:37 pm
I was actually a bit late this month making by shift because I just glanced at the UER instead of actually checking out the moving average, when I discovered that the UER crossed its 12 MMA by .01%. I've got reminders set up now to check this the day the BLS releases the data. Henceforth, I'll make whatever trades I need to on that day of the month.
This type of behavior changes the whole distribution and also shows that the process is not objective at all.

Cheers :greedy
How does it show that the process "is not objective at all?" Shifting the day of the month that the trades are made cannot change the results much when you consider that results were fairly stable from using the 3 MMA to over a 12 MMA. A strategy like this doesn't hinge on tiny implementation details.

And 'missing' the first day to trade by a week or so isn't going to significantly change "the whole distribution." Yes, I know that my results are going to vary from the backtests.

And while my first trade here was perhaps a bit "careless," it was not at all "subjective."
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

dknightd
Posts: 1872
Joined: Wed Mar 07, 2018 11:57 am

Re: My trend following strategy and experience

Post by dknightd » Sat Jan 19, 2019 1:16 pm

I think it is a good idea to have a plan and stick to it.

james22
Posts: 1453
Joined: Tue Aug 21, 2007 2:22 pm

Re: My trend following strategy and experience

Post by james22 » Sat Jan 19, 2019 2:40 pm

Interesting, but I'd always fear "it's different this time."

I've the greatest respect for Hussman's analysis, but he's been surprised by ahistoric QE.

I move between Stable Value, Small Value, and Emerging Market indices in my 401k based on valuation, but I've no specific measure/s. Rather I read quite a bit and move as I'm convinced.

Is this a Plan? It doesn't lend itself to an IPS, no, but given a complex, adaptive market, it works for me.

Post Reply