Tactical Asset Allocation + HBPP an intriguing combo

A place to talk about speculative investing ideas for the optional Variable Portfolio

Moderator: Global Moderator

User avatar
InsuranceGuy
Executive Member
Executive Member
Posts: 227
Joined: Sun Mar 29, 2015 1:44 pm

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by InsuranceGuy » Mon Dec 17, 2018 7:22 pm

ochotona wrote:
Mon Dec 17, 2018 3:46 pm
I can't wait to sell this binch on December 31
I'm just glad my model bailed at the end of November.
User avatar
ochotona
Executive Member
Executive Member
Posts: 2832
Joined: Fri Jan 30, 2015 5:54 am

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by ochotona » Mon Dec 17, 2018 8:05 pm

InsuranceGuy wrote:
Mon Dec 17, 2018 7:22 pm
ochotona wrote:
Mon Dec 17, 2018 3:46 pm
I can't wait to sell this binch on December 31
I'm just glad my model bailed at the end of November.
That's my price for having a less-twitchy model. Deeper draws.
HappyMan
Full Member
Full Member
Posts: 87
Joined: Thu Nov 09, 2017 7:01 am

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by HappyMan » Mon Dec 17, 2018 9:30 pm

With you, ochotona.
User avatar
InsuranceGuy
Executive Member
Executive Member
Posts: 227
Joined: Sun Mar 29, 2015 1:44 pm

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by InsuranceGuy » Mon Dec 17, 2018 9:34 pm

Yep, sometimes those extra trades help and other times not. GEM is solid, much of the time my model and GEM give the same or similar signals.
HappyMan
Full Member
Full Member
Posts: 87
Joined: Thu Nov 09, 2017 7:01 am

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by HappyMan » Mon Dec 17, 2018 9:49 pm

Would be cool to continue comparing them. Keep 'em coming, IG.
User avatar
Kbg
Executive Member
Executive Member
Posts: 1368
Joined: Fri May 23, 2014 4:18 pm

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by Kbg » Wed Dec 19, 2018 6:39 pm

I'm tweaking my dual mo stuff to the 12mo signal and a 6/3 mo weighted signal in 2019. 50% of the port to each signal. This is a small change to my current 10mo/5mo approach.
HappyMan
Full Member
Full Member
Posts: 87
Joined: Thu Nov 09, 2017 7:01 am

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by HappyMan » Wed Dec 19, 2018 10:17 pm

Kbg wrote:
Wed Dec 19, 2018 6:39 pm
I'm tweaking my dual mo stuff to the 12mo signal and a 6/3 mo weighted signal in 2019. 50% of the port to each signal. This is a small change to my current 10mo/5mo approach.
Interesting. Expectant results?
User avatar
Kbg
Executive Member
Executive Member
Posts: 1368
Joined: Fri May 23, 2014 4:18 pm

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by Kbg » Wed Dec 19, 2018 11:54 pm

HappyMan wrote:
Wed Dec 19, 2018 10:17 pm
Kbg wrote:
Wed Dec 19, 2018 6:39 pm
I'm tweaking my dual mo stuff to the 12mo signal and a 6/3 mo weighted signal in 2019. 50% of the port to each signal. This is a small change to my current 10mo/5mo approach.
Interesting. Expectant results?
No idea. They both backtest well with the ST clocking the 12mo one by a good stretch in more recent years. However, the main thing for me is that the odds are better for longer term success as I'm going to get the average of whatever they end up doing in the future. It would be nice to know which one is going to be best for the next 15 years, but no one does. Additionally, the look back periods perform differently based on market type so there is some system diversification benefit. The LT is best for most of the ride during long term bull markets normally, while the ST is better for choppier markets and absolutely kills it coming out of bear bottoms.
User avatar
InsuranceGuy
Executive Member
Executive Member
Posts: 227
Joined: Sun Mar 29, 2015 1:44 pm

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by InsuranceGuy » Thu Dec 20, 2018 8:39 pm

Kbg wrote:
Wed Dec 19, 2018 11:54 pm
No idea. They both backtest well with the ST clocking the 12mo one by a good stretch in more recent years. However, the main thing for me is that the odds are better for longer term success as I'm going to get the average of whatever they end up doing in the future. It would be nice to know which one is going to be best for the next 15 years, but no one does. Additionally, the look back periods perform differently based on market type so there is some system diversification benefit. The LT is best for most of the ride during long term bull markets normally, while the ST is better for choppier markets and absolutely kills it coming out of bear bottoms.
Have you thought about sticking with the longer 12 month with a volatility filter? The thing I don't like about the 3/6 month filter is that more gains are short term rather than long term.
User avatar
Kbg
Executive Member
Executive Member
Posts: 1368
Joined: Fri May 23, 2014 4:18 pm

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by Kbg » Fri Dec 21, 2018 6:58 am

IG,

Definite downside to ST is taxes. You will have to explain the volatility filter application. Using a ST look back and volatility is kinda the whole point of going shorter...ie take advantage of shorter term movements.
User avatar
InsuranceGuy
Executive Member
Executive Member
Posts: 227
Joined: Sun Mar 29, 2015 1:44 pm

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by InsuranceGuy » Fri Dec 21, 2018 9:51 am

Kbg -

I try to keep things simple so to filter on volatility I just look at the annualized standard deviation of the log returns for the last month (21 trading days) or two months (42 trading days). No matter the signal from the model if is above a certain threshold say 25%ish then either pull back to 50% invested or completely out. I'm pretty something similar was done in the AAA paper so you could look there or https://www.investing.com/analysis/comb ... nce-130378 shows a sensitivity test to various parameter selection.

Another thing I do to ensure long term gains is shorten from 12mo to 10mo and only change course when both the current 10mo and prior month 10mo give the same signal. For example if I am in S&P at the end of this month and at the end of the next month is International I would not switch but stay in the S&P. I would not switch until the next month if Internation is indicated after a 2nd future month if that makes sense.

I find this keeps the whipsaws and trades to a minimum, but at the same time I get divested when volatility jumps and things start to go south.

Hope that helps,
IG
User avatar
ochotona
Executive Member
Executive Member
Posts: 2832
Joined: Fri Jan 30, 2015 5:54 am

Re: Dual Momentum GEM + HBPP a great combo, easy to test

Post by ochotona » Sat Dec 22, 2018 6:46 pm

It truly sucks to get a sell signal when so deeply oversold. If we don't get a post-Christmas window dressing rally, I might not sell immediately on Dec 31 / Jan 1, I might wait for a bounce back towards 2600. I'm already only 40% equities so I'm not so deeply committed that I have to drop the hot potato.

I'll bet lots of tax loss harvest selling has taken place this year.
Post Reply