HEDGEFUNDIE's excellent adventure Part II: The next journey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
danyboy7
Posts: 198
Joined: Tue Jul 14, 2020 11:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by danyboy7 »

perfectuncertainty wrote: Mon Aug 31, 2020 12:41 pm
Sushmit wrote: Mon Aug 31, 2020 10:57 am
perfectuncertainty wrote: Mon Aug 31, 2020 10:39 am

I have a spreadsheet that does it.
Can you share it please, I am curious about it too
Link

I havent updated it - since I don't use TV. I use a differrent method using inverse volatility
Thanks a lot,how do we manage to update it ?
I have seen the light
User avatar
Meaty
Posts: 796
Joined: Mon Jul 22, 2013 7:35 pm
Location: Texas

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Meaty »

Stef wrote: Fri Sep 04, 2020 11:04 am Is TQQQ/TMF really a good idea?
HF himself didn’t think so. He made reference to alternatives to UPRO in his first post. The volatility is much greater than UPRO.
"Discipline equals Freedom" - Jocko Willink
User avatar
Meaty
Posts: 796
Joined: Mon Jul 22, 2013 7:35 pm
Location: Texas

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Meaty »

***deleted duplicate post
"Discipline equals Freedom" - Jocko Willink
vijaym73
Posts: 35
Joined: Fri Sep 08, 2017 11:34 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by vijaym73 »

If you can stomach the volatility then yes TQQQ/TMF can be used.
langlands
Posts: 569
Joined: Wed Apr 03, 2019 10:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by langlands »

RovenSkyfall wrote: Fri Sep 04, 2020 7:01 am M1 seems to only have SIPC insurance up to $500K.

Is anyone doing this through a Roth IRA with Fidelity? If so, are there any fees for housing it under Fidelity? Their Agreement and Disclosures are surprisingly unclear about fees. Also, is the HFEA available? I opened a Vanguard Roth IRA to invest in the HFEA but they dont allow leveraged etfs, so that was a bit of a bust. Just looking for the best place to open an account for the HFE in a Roth IRA. Thank you for your time.
I don't understand, what's wrong with Fidelity? Why do you think there are "fees for housing it under Fidelity?" They ask you to sign a "Designated Investments Agreement," which is just something you need to sign to get permission to buy leveraged ETFs. But I don't recall any mention of fees.
guyinlaw
Posts: 708
Joined: Wed Jul 03, 2019 9:54 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by guyinlaw »

vijaym73 wrote: Fri Sep 04, 2020 11:23 am If you can stomach the volatility then yes TQQQ/TMF can be used.
TQQQ is still under water if you had invested in 2000.

SP500 is more diversified than QQQ NASDAQ 100. UPRO better IMo
Time is your friend; impulse is your enemy. - John C. Bogle
jarjarM
Posts: 276
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

langlands wrote: Fri Sep 04, 2020 11:26 am
RovenSkyfall wrote: Fri Sep 04, 2020 7:01 am M1 seems to only have SIPC insurance up to $500K.

Is anyone doing this through a Roth IRA with Fidelity? If so, are there any fees for housing it under Fidelity? Their Agreement and Disclosures are surprisingly unclear about fees. Also, is the HFEA available? I opened a Vanguard Roth IRA to invest in the HFEA but they dont allow leveraged etfs, so that was a bit of a bust. Just looking for the best place to open an account for the HFE in a Roth IRA. Thank you for your time.
I don't understand, what's wrong with Fidelity? Why do you think there are "fees for housing it under Fidelity?" They ask you to sign a "Designated Investments Agreement," which is just something you need to sign to get permission to buy leveraged ETFs. But I don't recall any mention of fees.
I run this at Fidelity. No issue and fee there other than signing that designated investment agreement (you do pay exchange fee for trade, but that's true anywhere).
langlands
Posts: 569
Joined: Wed Apr 03, 2019 10:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by langlands »

guyinlaw wrote: Fri Sep 04, 2020 11:54 am
vijaym73 wrote: Fri Sep 04, 2020 11:23 am If you can stomach the volatility then yes TQQQ/TMF can be used.
TQQQ is still under water if you had invested in 2000.

SP500 is more diversified than QQQ NASDAQ 100. UPRO better IMo
That is really not the right way to think about it.

Everyone here (well hopefully) understands that neither TQQQ nor UPRO is a set it and forget it investment. Portfolio allocation is crucial for successfully using either product. (The providers just lazily say it's only appropriate for daily use, but that's just because explaining in detail how it could be used for long term use would be way too complicated. Exhibit A: Two 100+ page threads on this forum) One way to describe why it works is "volatility pumping."
Last edited by langlands on Fri Sep 04, 2020 12:08 pm, edited 1 time in total.
langlands
Posts: 569
Joined: Wed Apr 03, 2019 10:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by langlands »

jarjarM wrote: Fri Sep 04, 2020 11:58 am
langlands wrote: Fri Sep 04, 2020 11:26 am
RovenSkyfall wrote: Fri Sep 04, 2020 7:01 am M1 seems to only have SIPC insurance up to $500K.

Is anyone doing this through a Roth IRA with Fidelity? If so, are there any fees for housing it under Fidelity? Their Agreement and Disclosures are surprisingly unclear about fees. Also, is the HFEA available? I opened a Vanguard Roth IRA to invest in the HFEA but they dont allow leveraged etfs, so that was a bit of a bust. Just looking for the best place to open an account for the HFE in a Roth IRA. Thank you for your time.
I don't understand, what's wrong with Fidelity? Why do you think there are "fees for housing it under Fidelity?" They ask you to sign a "Designated Investments Agreement," which is just something you need to sign to get permission to buy leveraged ETFs. But I don't recall any mention of fees.
I run this at Fidelity. No issue and fee there other than signing that designated investment agreement (you do pay exchange fee for trade, but that's true anywhere).
Thanks. What do you mean by exchange fee? That's not anything specific to leveraged funds right? You just mean the extremely nominal fee on the order of a few cents per order (approximately 0.001-0.01% cost) for any stock or ETF trade right?
jarjarM
Posts: 276
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

langlands wrote: Fri Sep 04, 2020 12:06 pm
jarjarM wrote: Fri Sep 04, 2020 11:58 am
langlands wrote: Fri Sep 04, 2020 11:26 am
RovenSkyfall wrote: Fri Sep 04, 2020 7:01 am M1 seems to only have SIPC insurance up to $500K.

Is anyone doing this through a Roth IRA with Fidelity? If so, are there any fees for housing it under Fidelity? Their Agreement and Disclosures are surprisingly unclear about fees. Also, is the HFEA available? I opened a Vanguard Roth IRA to invest in the HFEA but they dont allow leveraged etfs, so that was a bit of a bust. Just looking for the best place to open an account for the HFE in a Roth IRA. Thank you for your time.
I don't understand, what's wrong with Fidelity? Why do you think there are "fees for housing it under Fidelity?" They ask you to sign a "Designated Investments Agreement," which is just something you need to sign to get permission to buy leveraged ETFs. But I don't recall any mention of fees.
I run this at Fidelity. No issue and fee there other than signing that designated investment agreement (you do pay exchange fee for trade, but that's true anywhere).
Thanks. What do you mean by exchange fee? That's not anything specific to leveraged funds right? You just mean the extremely nominal fee on the order of a few cents per order (approximately 0.001-0.01% cost) for any stock or ETF trade right?
Yup, that's what I'm talking about, the nominal fee per order (pennies mostly)
User avatar
cos
Posts: 223
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by cos »

langlands wrote: Fri Sep 04, 2020 12:06 pm You just mean the extremely nominal fee on the order of a few cents per order (approximately 0.001-0.01% cost) for any stock or ETF trade right?
We're all talking about the Section 31 Transaction Fee enforced by the SEC, right? On top of being microscopic, isn't that fee only for sell orders?
BogleBobby
Posts: 38
Joined: Tue Jul 30, 2019 9:56 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by BogleBobby »

danyboy7 wrote: Fri Sep 04, 2020 11:03 am
tomphilly wrote: Fri Sep 04, 2020 8:38 am
danyboy7 wrote: Fri Sep 04, 2020 5:07 am The target 25% volatility strategy for UPRO+TMF combo seems to be the best Hedgeundie version right ? Do I have to get the PV pro version to see the signals change allocation,right ?
It's not necessarily better - it has much higher drawdowns and lower Sharpe than classic HFEA - at 80/20 35% volatility it has a superior CAGR, at least based on 10 year backtests and a few important toggle settings. It also hasn't been backtested beyond 2010, unlike HFEA which was modeled back to the 1930's by HedgeFundie. I would love if a smart math/finance person on here modeled it during the periods where HFEA under-performed (60's-80's). My motivation to move to it was not based on higher CAGR, but its reduced reliance on treasuries. You need the PV subscription to see the forward signal, though there may be ways to calculate it manually.
danyboy7 wrote: Fri Sep 04, 2020 5:07 am Is there anyone that is performing HF strategies through options ?
I do a small allocation of RSP calls (equal weight SPY), my overall default allocation is 10/70/20 RSP/UPRO/TMF. My gambit is RSP will begin to outperform SPY in the near term. The spread on calls is large, so it's not ideal, so I've kept the experiment small - if there was a 3xRSP I'd use that instead.

This morning the PV volatility signal is calling for 33% TMF, so I have re-balanced... reluctantly - TMF is dumping again.
Thanks a lot,yes a 10 year backtest is no enough imho,we need to go much further back
The daily simulated dataset goes back to 1987, so you can backtest the adaptive allocation methods over a period of ~33 years. From what I can remember, the adaptive allocation methods perform better than fixed allocations if you go back to 1987.

Also, the further you go back in time, the less relevant the date becomes. Our financial systems were different back then, including how we managed interest rates, how the government responded to times of crisis, etc.

Due to changing government policy, I'd be hesitant to read too much into backtests which go far back into history (like the 60's).
User avatar
danyboy7
Posts: 198
Joined: Tue Jul 14, 2020 11:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by danyboy7 »

BogleBobby wrote: Fri Sep 04, 2020 12:28 pm
danyboy7 wrote: Fri Sep 04, 2020 11:03 am
tomphilly wrote: Fri Sep 04, 2020 8:38 am
danyboy7 wrote: Fri Sep 04, 2020 5:07 am The target 25% volatility strategy for UPRO+TMF combo seems to be the best Hedgeundie version right ? Do I have to get the PV pro version to see the signals change allocation,right ?
It's not necessarily better - it has much higher drawdowns and lower Sharpe than classic HFEA - at 80/20 35% volatility it has a superior CAGR, at least based on 10 year backtests and a few important toggle settings. It also hasn't been backtested beyond 2010, unlike HFEA which was modeled back to the 1930's by HedgeFundie. I would love if a smart math/finance person on here modeled it during the periods where HFEA under-performed (60's-80's). My motivation to move to it was not based on higher CAGR, but its reduced reliance on treasuries. You need the PV subscription to see the forward signal, though there may be ways to calculate it manually.
danyboy7 wrote: Fri Sep 04, 2020 5:07 am Is there anyone that is performing HF strategies through options ?
I do a small allocation of RSP calls (equal weight SPY), my overall default allocation is 10/70/20 RSP/UPRO/TMF. My gambit is RSP will begin to outperform SPY in the near term. The spread on calls is large, so it's not ideal, so I've kept the experiment small - if there was a 3xRSP I'd use that instead.

This morning the PV volatility signal is calling for 33% TMF, so I have re-balanced... reluctantly - TMF is dumping again.
Thanks a lot,yes a 10 year backtest is no enough imho,we need to go much further back
The daily simulated dataset goes back to 1987, so you can backtest the adaptive allocation methods over a period of ~33 years. From what I can remember, the adaptive allocation methods perform better than fixed allocations if you go back to 1987.

Also, the further you go back in time, the less relevant the date becomes. Our financial systems were different back then, including how we managed interest rates, how the government responded to times of crisis, etc.

Due to changing government policy, I'd be hesitant to read too much into backtests which go far back into history (like the 60's).
No I can't because in the optimization portfolio section,PV doesn't allow you to use negative cashX to simulate leveraged ETFs.So you're forced to use upro+tmf that can can go back only to 2009
I have seen the light
BogleBobby
Posts: 38
Joined: Tue Jul 30, 2019 9:56 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by BogleBobby »

I continue to struggle with the exact method to implement for this strategy because I want to reduce my reliance on the TMF due to the low interest rate environment.

I've been implementing an adaptive allocation method and plan to stay the course here.

But have been fidgeting with minimums and maximums on the UPRO portion.

I am considering having a minimum of a 55% allocation to UPRO. This restriction appears to reduce return in backtests, but would ensure that I'm never too heavily allocated to TMF.
tomphilly
Posts: 59
Joined: Wed Aug 05, 2020 11:29 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by tomphilly »

danyboy7 wrote: Fri Sep 04, 2020 12:32 pm in the optimization portfolio section,PV doesn't allow you to use negative cashX to simulate leveraged ETFs.So you're forced to use upro+tmf that can can go back only to 2009
Yup, it's same with the target volatility modeling - you can't use negative CASHX. The closest thing you have is ULPIX (2x SP500) which goes back to 1997, however there's no corresponding 2x leveraged LTT of a similar age that I know of.
BogleBobby
Posts: 38
Joined: Tue Jul 30, 2019 9:56 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by BogleBobby »

danyboy7 wrote: Fri Sep 04, 2020 12:32 pm
BogleBobby wrote: Fri Sep 04, 2020 12:28 pm
danyboy7 wrote: Fri Sep 04, 2020 11:03 am
tomphilly wrote: Fri Sep 04, 2020 8:38 am
danyboy7 wrote: Fri Sep 04, 2020 5:07 am The target 25% volatility strategy for UPRO+TMF combo seems to be the best Hedgeundie version right ? Do I have to get the PV pro version to see the signals change allocation,right ?
It's not necessarily better - it has much higher drawdowns and lower Sharpe than classic HFEA - at 80/20 35% volatility it has a superior CAGR, at least based on 10 year backtests and a few important toggle settings. It also hasn't been backtested beyond 2010, unlike HFEA which was modeled back to the 1930's by HedgeFundie. I would love if a smart math/finance person on here modeled it during the periods where HFEA under-performed (60's-80's). My motivation to move to it was not based on higher CAGR, but its reduced reliance on treasuries. You need the PV subscription to see the forward signal, though there may be ways to calculate it manually.
danyboy7 wrote: Fri Sep 04, 2020 5:07 am Is there anyone that is performing HF strategies through options ?
I do a small allocation of RSP calls (equal weight SPY), my overall default allocation is 10/70/20 RSP/UPRO/TMF. My gambit is RSP will begin to outperform SPY in the near term. The spread on calls is large, so it's not ideal, so I've kept the experiment small - if there was a 3xRSP I'd use that instead.

This morning the PV volatility signal is calling for 33% TMF, so I have re-balanced... reluctantly - TMF is dumping again.
Thanks a lot,yes a 10 year backtest is no enough imho,we need to go much further back
The daily simulated dataset goes back to 1987, so you can backtest the adaptive allocation methods over a period of ~33 years. From what I can remember, the adaptive allocation methods perform better than fixed allocations if you go back to 1987.

Also, the further you go back in time, the less relevant the date becomes. Our financial systems were different back then, including how we managed interest rates, how the government responded to times of crisis, etc.

Due to changing government policy, I'd be hesitant to read too much into backtests which go far back into history (like the 60's).
No I can't because in the optimization portfolio section,PV doesn't allow you to use negative cashX to simulate leveraged ETFs.So you're forced to use upro+tmf that can can go back only to 2009
There's a daily dataset that's linked in the original post that goes back to 1986. You can upload this to PV and backtest to 1986.

See here.
User avatar
danyboy7
Posts: 198
Joined: Tue Jul 14, 2020 11:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by danyboy7 »

BogleBobby wrote: Fri Sep 04, 2020 12:42 pm
danyboy7 wrote: Fri Sep 04, 2020 12:32 pm
BogleBobby wrote: Fri Sep 04, 2020 12:28 pm
danyboy7 wrote: Fri Sep 04, 2020 11:03 am
tomphilly wrote: Fri Sep 04, 2020 8:38 am

It's not necessarily better - it has much higher drawdowns and lower Sharpe than classic HFEA - at 80/20 35% volatility it has a superior CAGR, at least based on 10 year backtests and a few important toggle settings. It also hasn't been backtested beyond 2010, unlike HFEA which was modeled back to the 1930's by HedgeFundie. I would love if a smart math/finance person on here modeled it during the periods where HFEA under-performed (60's-80's). My motivation to move to it was not based on higher CAGR, but its reduced reliance on treasuries. You need the PV subscription to see the forward signal, though there may be ways to calculate it manually.



I do a small allocation of RSP calls (equal weight SPY), my overall default allocation is 10/70/20 RSP/UPRO/TMF. My gambit is RSP will begin to outperform SPY in the near term. The spread on calls is large, so it's not ideal, so I've kept the experiment small - if there was a 3xRSP I'd use that instead.

This morning the PV volatility signal is calling for 33% TMF, so I have re-balanced... reluctantly - TMF is dumping again.
Thanks a lot,yes a 10 year backtest is no enough imho,we need to go much further back
The daily simulated dataset goes back to 1987, so you can backtest the adaptive allocation methods over a period of ~33 years. From what I can remember, the adaptive allocation methods perform better than fixed allocations if you go back to 1987.

Also, the further you go back in time, the less relevant the date becomes. Our financial systems were different back then, including how we managed interest rates, how the government responded to times of crisis, etc.

Due to changing government policy, I'd be hesitant to read too much into backtests which go far back into history (like the 60's).
No I can't because in the optimization portfolio section,PV doesn't allow you to use negative cashX to simulate leveraged ETFs.So you're forced to use upro+tmf that can can go back only to 2009
There's a daily dataset that's linked in the original post that goes back to 1986. You can upload this to PV and backtest to 1986.

See here.
Thanks a lot,how do I upload them on PV ? Is it required the Pro-version ?
I have seen the light
langlands
Posts: 569
Joined: Wed Apr 03, 2019 10:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by langlands »

There's not some edict that you can only use UPRO and TMF. There's also something called cash. There seems to be some misconception that there is something magical about the HFEA. It's just a leveraged stock/bond portfolio. If you don't like bonds right now, there's always the option to leverage only stocks, i.e. UPRO and cash. IMO, the most important number to keep track of is how much equity leverage you're using. If you think along these lines, you realize there's a lot of connections to lifecycle investing.
jarjarM
Posts: 276
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

danyboy7 wrote: Fri Sep 04, 2020 12:48 pm
Thanks a lot,how do I upload them on PV ? Is it required the Pro-version ?
It's being done. I did this as a pet project during the shelter at home fun.

Image

Code: Select all

		UPRO Baseline	55/45 Portfolio	VolatilityPortfolio	MomentumPortfolio	AdaptiveAssetAllocation
cagr		10.0			14.9		17.8			15.0			18.1
sharpe		0.46			0.62		0.80			0.64			0.82
volatility	0.54			0.29		0.24			0.28			0.24
max_drawdown	-98.2			-74.5		-44.2			-69.9			-39.2
60/40 vs. 55/45 didn't show much difference so I didn't simulate it separately. The data is to 2/2019 when the simulated data end. Volatility is using inverse volatility from prior month, momentum is best performing part from prior 6 months. Adaptive AA is minimum variance. Have fun.
RovenSkyfall
Posts: 82
Joined: Wed Apr 01, 2020 11:40 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by RovenSkyfall »

How does the Adaptive Asset Allocation differ from the volatility?
tomphilly
Posts: 59
Joined: Wed Aug 05, 2020 11:29 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by tomphilly »

jarjarM wrote: Fri Sep 04, 2020 1:12 pm Image
Thanks for this. What is the default ratio and the target volatility of the VolatilityPortfolio here? Also, is it UPRO/TMF or something else?
Hydromod
Posts: 419
Joined: Tue Mar 26, 2019 10:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Hydromod »

danyboy7 wrote: Fri Sep 04, 2020 12:48 pm Thanks a lot,how do I upload them on PV ? Is it required the Pro-version ?
You used to be able to freely upload data, but now it requires the subscription.
Mickelous
Posts: 132
Joined: Mon Apr 20, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Mickelous »

Is someone willing to summarize the adaptive allocation model? How often would I have to make changes to that, once a month? Just making sure I don't have to check VIX every day or something.
jarjarM
Posts: 276
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

Mickelous wrote: Fri Sep 04, 2020 6:08 pm Is someone willing to summarize the adaptive allocation model? How often would I have to make changes to that, once a month? Just making sure I don't have to check VIX every day or something.
Here is the link to the strategy from smartly.

https://allocatesmartly.com/adam-butler ... llocation/

Strategy rules tested is based on smartly's method:
- At the close on the last trading day of the month, calculate the 6-month return for UPRO/TMF.
- Weight the 2 assets according to minimum variance optimization, using a “weighted” covariance matrix calculated based on 6 month correlation and 1 month (21 trading day) volatility.
- Hold positions until the final trading day of the following month. Rebalance the entire portfolio monthly, regardless of whether there is a change in position.
jarjarM
Posts: 276
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

tomphilly wrote: Fri Sep 04, 2020 1:30 pm

Thanks for this. What is the default ratio and the target volatility of the VolatilityPortfolio here? Also, is it UPRO/TMF or something else?
The volatility shown in graph is using inverse volatility to determine the ratio of the 2 assets, it's not targeted volatility. Though it should be fairly easy to program it. What kind of volatility target are you looking for, 30%? Also, what would be your out of market mix (TMF only or UPRO/TMF mix)?
Mickelous
Posts: 132
Joined: Mon Apr 20, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Mickelous »

jarjarM wrote: Fri Sep 04, 2020 8:03 pm
Mickelous wrote: Fri Sep 04, 2020 6:08 pm Is someone willing to summarize the adaptive allocation model? How often would I have to make changes to that, once a month? Just making sure I don't have to check VIX every day or something.
Here is the link to the strategy from smartly.

https://allocatesmartly.com/adam-butler ... llocation/

Strategy rules tested is based on smartly's method:
- Weight the 2 assets according to minimum variance optimization, using a “weighted” covariance matrix calculated based on 6 month correlation and 1 month (21 trading day) volatility.
Is there a more thorough explanation of this?

How do I weigh assets according to minimum variance optimization? What is a weighted covariance matrix and how do I use it?

edit: So I was able to use portfoliovisualizer for the minimum variance but it made me pick a 1 year period, wouldn't go to 6 months. Any tool that will let me choose a 6 month time frame so I don't have to use a complex formula?

EDIT: I'm understanding this a bit more, I just can't find a easy to to pull all the data and try to put everything together with a formula on a monthly basis, or if there is a free tool online that can pull daily data for tickets and come up with an allocation for given time period. I have found a paid tool but there must be a way that I can do this for free and not take up much time. Even the excel route would require me to download the daily data from another website and import that into a spreadsheet. I am having a tough time wrapping my head around the formula that I have found to calculate the weight of two assets using the minimum variance asset allocation formula using volatility.
tomphilly
Posts: 59
Joined: Wed Aug 05, 2020 11:29 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by tomphilly »

jarjarM wrote: Fri Sep 04, 2020 8:07 pm
tomphilly wrote: Fri Sep 04, 2020 1:30 pm

Thanks for this. What is the default ratio and the target volatility of the VolatilityPortfolio here? Also, is it UPRO/TMF or something else?
The volatility shown in graph is using inverse volatility to determine the ratio of the 2 assets, it's not targeted volatility. Though it should be fairly easy to program it. What kind of volatility target are you looking for, 30%? Also, what would be your out of market mix (TMF only or UPRO/TMF mix)?
Right now I'm doing a default 80/20 TMF at 35% target volatility (downside volatility only). As of yesterday's selloff the portfolio visualizer forward ratio signal has changed to 67/33 UPRO/TMF. Does this differ much from your strategy?
Saaybogle
Posts: 24
Joined: Tue Jun 30, 2015 8:42 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Saaybogle »

Mickelous wrote: Fri Sep 04, 2020 9:16 pm
jarjarM wrote: Fri Sep 04, 2020 8:03 pm
Mickelous wrote: Fri Sep 04, 2020 6:08 pm Is someone willing to summarize the adaptive allocation model? How often would I have to make changes to that, once a month? Just making sure I don't have to check VIX every day or something.
Here is the link to the strategy from smartly.

https://allocatesmartly.com/adam-butler ... llocation/

Strategy rules tested is based on smartly's method:
- Weight the 2 assets according to minimum variance optimization, using a “weighted” covariance matrix calculated based on 6 month correlation and 1 month (21 trading day) volatility.
Is there a more thorough explanation of this?

How do I weigh assets according to minimum variance optimization? What is a weighted covariance matrix and how do I use it?

edit: So I was able to use portfoliovisualizer for the minimum variance but it made me pick a 1 year period, wouldn't go to 6 months. Any tool that will let me choose a 6 month time frame so I don't have to use a complex formula?

EDIT: I'm understanding this a bit more, I just can't find a easy to to pull all the data and try to put everything together with a formula on a monthly basis, or if there is a free tool online that can pull daily data for tickets and come up with an allocation for given time period. I have found a paid tool but there must be a way that I can do this for free and not take up much time. Even the excel route would require me to download the daily data from another website and import that into a spreadsheet. I am having a tough time wrapping my head around the formula that I have found to calculate the weight of two assets using the minimum variance asset allocation formula using volatility.
I would love to hear about this as well. I understand adaptive allocation but would want to know if there is a simple, time-efficient way to figure out what the monthly re-allocation between UPRO and TMF should be.
BogleBobby
Posts: 38
Joined: Tue Jul 30, 2019 9:56 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by BogleBobby »

You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.
Mickelous
Posts: 132
Joined: Mon Apr 20, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Mickelous »

BogleBobby wrote: Sat Sep 05, 2020 6:29 pm You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.
It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
jarjarM
Posts: 276
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

Mickelous wrote: Sat Sep 05, 2020 8:54 pm
BogleBobby wrote: Sat Sep 05, 2020 6:29 pm You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.
It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
perfectuncertainty
Posts: 126
Joined: Sun Feb 04, 2018 7:44 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by perfectuncertainty »

jarjarM wrote: Sat Sep 05, 2020 9:05 pm
Mickelous wrote: Sat Sep 05, 2020 8:54 pm
BogleBobby wrote: Sat Sep 05, 2020 6:29 pm You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.
It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
Saaybogle
Posts: 24
Joined: Tue Jun 30, 2015 8:42 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Saaybogle »

perfectuncertainty wrote: Sat Sep 05, 2020 10:12 pm
jarjarM wrote: Sat Sep 05, 2020 9:05 pm
Mickelous wrote: Sat Sep 05, 2020 8:54 pm
BogleBobby wrote: Sat Sep 05, 2020 6:29 pm You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.

It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
This is great. Thanks. I noticed you use 22 days of data in the spreadsheet instead of 21. Any particular reason for that?
perfectuncertainty
Posts: 126
Joined: Sun Feb 04, 2018 7:44 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by perfectuncertainty »

The daily returns need 22 days else you can’t calculate a daily return for the first day.

The volatility and correlation coefficient formulas use 21 days.
User avatar
kerstverlichting
Posts: 27
Joined: Tue Feb 11, 2020 5:26 am
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by kerstverlichting »

perfectuncertainty wrote: Sat Sep 05, 2020 10:12 pm
jarjarM wrote: Sat Sep 05, 2020 9:05 pm
Mickelous wrote: Sat Sep 05, 2020 8:54 pm
BogleBobby wrote: Sat Sep 05, 2020 6:29 pm You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.
It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
Nice work. I quickly created an automated variant for convenience. It pulls the latest data from Yahoo and should auto refresh upon opening the file (if not, go to Data>Refresh all). I didn't mess with any of your calculations, though I added an input field for portfolio value that it will use to calculate the rebalancing requirements for you. As another bonus, I added drop down Stocks and Bonds fields, which you can use to switch between UPRO and TQQQ. The field also works for Bonds, but only TMF is available so it serves no real use (I only added the functionality in case some people might want to add other tickers themselves to compare).
https://ufile.io/csocipdg
EDIT: New version here: viewtopic.php?f=10&t=288192&p=5486976#p5486976

Maybe someone can host it on Google Sheets (I have no experience working with it) as this link will expire at some point.

Not sure it'll keep on working correctly once there's a stock split, at least not if a split happened in the past month. I can set Yahoo to use values adjusted for splits, but I don't know how such data would look like and thus I didn't use it.

Image
Last edited by kerstverlichting on Thu Sep 10, 2020 5:50 am, edited 3 times in total.
User avatar
danyboy7
Posts: 198
Joined: Tue Jul 14, 2020 11:35 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by danyboy7 »

kerstverlichting wrote: Mon Sep 07, 2020 6:45 am
perfectuncertainty wrote: Sat Sep 05, 2020 10:12 pm
jarjarM wrote: Sat Sep 05, 2020 9:05 pm
Mickelous wrote: Sat Sep 05, 2020 8:54 pm
BogleBobby wrote: Sat Sep 05, 2020 6:29 pm You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.
It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
Nice work. I quickly created an automated variant for convenience. It pulls the latest data from Yahoo and should auto refresh upon opening the file (if not, go to Data>Refresh all). I didn't mess with any of your calculations, though I added an input field for portfolio value that it will use to calculate the rebalancing requirements for you. As another bonus, I added drop down Stocks and Bonds fields, which you can use to switch between UPRO and TQQQ. The field also works for Bonds, but only TMF is available so it serves no real use (I only added the functionality in case some people might want to add other tickers themselves to compare).
https://ufile.io/csocipdg

Maybe someone can host it on Google Sheets (I have no experience working with it) as this link will expire at some point.

Not sure it'll keep on working correctly once there's a stock split, at least not if a split happened in the past month. I can set Yahoo to use values adjusted for splits, but I don't know how such data would look like and thus I didn't use it.

Image
Awesome,thanks a lot :beer :beer
I have seen the light
User avatar
kerstverlichting
Posts: 27
Joined: Tue Feb 11, 2020 5:26 am
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by kerstverlichting »

btw, what's with the talk about using 6 mos worth of data
- Weight the 2 assets according to minimum variance optimization, using a “weighted” covariance matrix calculated based on 6 month correlation and 1 month (21 trading day) volatility.
It's not used in this calculation. Should it be added?
Mickelous
Posts: 132
Joined: Mon Apr 20, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Mickelous »

kerstverlichting wrote: Mon Sep 07, 2020 6:45 am
perfectuncertainty wrote: Sat Sep 05, 2020 10:12 pm
jarjarM wrote: Sat Sep 05, 2020 9:05 pm
Mickelous wrote: Sat Sep 05, 2020 8:54 pm
BogleBobby wrote: Sat Sep 05, 2020 6:29 pm You can pull daily data from Yahoo finance and paste it into a spreadsheet. It's a slightly manual process, but if you're just pulling two tickers, it's not that much work to do each month. No need to pay for PV if you're only using it for the allocation.
It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
Nice work. I quickly created an automated variant for convenience. It pulls the latest data from Yahoo and should auto refresh upon opening the file (if not, go to Data>Refresh all). I didn't mess with any of your calculations, though I added an input field for portfolio value that it will use to calculate the rebalancing requirements for you. As another bonus, I added drop down Stocks and Bonds fields, which you can use to switch between UPRO and TQQQ. The field also works for Bonds, but only TMF is available so it serves no real use (I only added the functionality in case some people might want to add other tickers themselves to compare).
https://ufile.io/csocipdg

Maybe someone can host it on Google Sheets (I have no experience working with it) as this link will expire at some point.

Not sure it'll keep on working correctly once there's a stock split, at least not if a split happened in the past month. I can set Yahoo to use values adjusted for splits, but I don't know how such data would look like and thus I didn't use it.

Image
How hard would it be to add tqqq to this mix? Having a third asset for some us running the adventure with the tmf upro tqqq split
User avatar
kerstverlichting
Posts: 27
Joined: Tue Feb 11, 2020 5:26 am
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by kerstverlichting »

Mickelous wrote: Mon Sep 07, 2020 8:57 am
kerstverlichting wrote: Mon Sep 07, 2020 6:45 am
perfectuncertainty wrote: Sat Sep 05, 2020 10:12 pm
jarjarM wrote: Sat Sep 05, 2020 9:05 pm
Mickelous wrote: Sat Sep 05, 2020 8:54 pm

It's also how do you get the variables you need for the formula and an easy way to process the formula to figure out the weighing process for someone who isn't into advanced statistics.
Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
Nice work. I quickly created an automated variant for convenience. It pulls the latest data from Yahoo and should auto refresh upon opening the file (if not, go to Data>Refresh all). I didn't mess with any of your calculations, though I added an input field for portfolio value that it will use to calculate the rebalancing requirements for you. As another bonus, I added drop down Stocks and Bonds fields, which you can use to switch between UPRO and TQQQ. The field also works for Bonds, but only TMF is available so it serves no real use (I only added the functionality in case some people might want to add other tickers themselves to compare).
https://ufile.io/csocipdg

Maybe someone can host it on Google Sheets (I have no experience working with it) as this link will expire at some point.

Not sure it'll keep on working correctly once there's a stock split, at least not if a split happened in the past month. I can set Yahoo to use values adjusted for splits, but I don't know how such data would look like and thus I didn't use it.

Image
How hard would it be to add tqqq to this mix? Having a third asset for some us running the adventure with the tmf upro tqqq split
You can select TQQQ from the drop down to replace UPRO. However, there is no option for a thee way portfolio TMF/UPRO/TQQQ. Like I said, I didn't really mess with the calculations, just added some automation and a few features for convenience. But I suspect it wouldn't be too hard to alter the calculations to account for a third asset. I might have a look later this week.

Btw, can anyone provide a quick explanation of the difference between minimum variance and inverse volatility? Why use one over the other?
Mickelous
Posts: 132
Joined: Mon Apr 20, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Mickelous »

kerstverlichting wrote: Mon Sep 07, 2020 9:24 am
Mickelous wrote: Mon Sep 07, 2020 8:57 am
kerstverlichting wrote: Mon Sep 07, 2020 6:45 am
perfectuncertainty wrote: Sat Sep 05, 2020 10:12 pm
jarjarM wrote: Sat Sep 05, 2020 9:05 pm

Somewhere in the 100+ pages there were a discussion and sample formula shown. This fairly easy to do for 2 assets. I’m not on my computer right now but when I get a chance I’ll post it.

Edit: I think page 91 of the thread has it.
Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
Nice work. I quickly created an automated variant for convenience. It pulls the latest data from Yahoo and should auto refresh upon opening the file (if not, go to Data>Refresh all). I didn't mess with any of your calculations, though I added an input field for portfolio value that it will use to calculate the rebalancing requirements for you. As another bonus, I added drop down Stocks and Bonds fields, which you can use to switch between UPRO and TQQQ. The field also works for Bonds, but only TMF is available so it serves no real use (I only added the functionality in case some people might want to add other tickers themselves to compare).
https://ufile.io/csocipdg

Maybe someone can host it on Google Sheets (I have no experience working with it) as this link will expire at some point.

Not sure it'll keep on working correctly once there's a stock split, at least not if a split happened in the past month. I can set Yahoo to use values adjusted for splits, but I don't know how such data would look like and thus I didn't use it.

Image
How hard would it be to add tqqq to this mix? Having a third asset for some us running the adventure with the tmf upro tqqq split
You can select TQQQ from the drop down to replace UPRO. However, there is no option for a thee way portfolio TMF/UPRO/TQQQ. Like I said, I didn't really mess with the calculations, just added some automation and a few features for convenience. But I suspect it wouldn't be too hard to alter the calculations to account for a third asset. I might have a look later this week.

Btw, can anyone provide a quick explanation of the difference between minimum variance and inverse volatility? Why use one over the other?
I got a trial for a website that does the calculations and shows the difference with a monte carlo sim and volatility portfolio did a little better.
Perfect Uncertainty
Posts: 12
Joined: Wed Jun 24, 2020 10:20 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Perfect Uncertainty »

kerstverlichting wrote: Mon Sep 07, 2020 8:52 am btw, what's with the talk about using 6 mos worth of data
- Weight the 2 assets according to minimum variance optimization, using a “weighted” covariance matrix calculated based on 6 month correlation and 1 month (21 trading day) volatility.
It's not used in this calculation. Should it be added?
The calculations obtain the exact same AAA values as PV.

I use Excel on a Mac so maybe that’s an issue but the 3 queries you have error out. Maybe there is something you didn’t include in your version?
Perfect Uncertainty
Posts: 12
Joined: Wed Jun 24, 2020 10:20 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Perfect Uncertainty »

You could use a 6 month covariance - it depends on which AAA sub model you are using on PV and which inputs.
Perfect Uncertainty
Posts: 12
Joined: Wed Jun 24, 2020 10:20 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Perfect Uncertainty »

Mickelous wrote: Mon Sep 07, 2020 9:46 am
kerstverlichting wrote: Mon Sep 07, 2020 9:24 am
Mickelous wrote: Mon Sep 07, 2020 8:57 am
kerstverlichting wrote: Mon Sep 07, 2020 6:45 am
perfectuncertainty wrote: Sat Sep 05, 2020 10:12 pm

Here is a spreadsheet I created that shows AAA between UPRO and TMF for both Inverse Volatility and Minimum Variance.

AAA Sample
Nice work. I quickly created an automated variant for convenience. It pulls the latest data from Yahoo and should auto refresh upon opening the file (if not, go to Data>Refresh all). I didn't mess with any of your calculations, though I added an input field for portfolio value that it will use to calculate the rebalancing requirements for you. As another bonus, I added drop down Stocks and Bonds fields, which you can use to switch between UPRO and TQQQ. The field also works for Bonds, but only TMF is available so it serves no real use (I only added the functionality in case some people might want to add other tickers themselves to compare).
https://ufile.io/csocipdg

Maybe someone can host it on Google Sheets (I have no experience working with it) as this link will expire at some point.

Not sure it'll keep on working correctly once there's a stock split, at least not if a split happened in the past month. I can set Yahoo to use values adjusted for splits, but I don't know how such data would look like and thus I didn't use it.

Image
How hard would it be to add tqqq to this mix? Having a third asset for some us running the adventure with the tmf upro tqqq split
You can select TQQQ from the drop down to replace UPRO. However, there is no option for a thee way portfolio TMF/UPRO/TQQQ. Like I said, I didn't really mess with the calculations, just added some automation and a few features for convenience. But I suspect it wouldn't be too hard to alter the calculations to account for a third asset. I might have a look later this week.

Btw, can anyone provide a quick explanation of the difference between minimum variance and inverse volatility? Why use one over the other?
I got a trial for a website that does the calculations and shows the difference with a monte carlo sim and volatility portfolio did a little better.
What tickers? What period? What website?
Mickelous
Posts: 132
Joined: Mon Apr 20, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Mickelous »

Perfect Uncertainty wrote: Mon Sep 07, 2020 10:43 am
Mickelous wrote: Mon Sep 07, 2020 9:46 am
kerstverlichting wrote: Mon Sep 07, 2020 9:24 am
Mickelous wrote: Mon Sep 07, 2020 8:57 am
kerstverlichting wrote: Mon Sep 07, 2020 6:45 am

Nice work. I quickly created an automated variant for convenience. It pulls the latest data from Yahoo and should auto refresh upon opening the file (if not, go to Data>Refresh all). I didn't mess with any of your calculations, though I added an input field for portfolio value that it will use to calculate the rebalancing requirements for you. As another bonus, I added drop down Stocks and Bonds fields, which you can use to switch between UPRO and TQQQ. The field also works for Bonds, but only TMF is available so it serves no real use (I only added the functionality in case some people might want to add other tickers themselves to compare).
https://ufile.io/csocipdg

Maybe someone can host it on Google Sheets (I have no experience working with it) as this link will expire at some point.

Not sure it'll keep on working correctly once there's a stock split, at least not if a split happened in the past month. I can set Yahoo to use values adjusted for splits, but I don't know how such data would look like and thus I didn't use it.

Image
How hard would it be to add tqqq to this mix? Having a third asset for some us running the adventure with the tmf upro tqqq split
You can select TQQQ from the drop down to replace UPRO. However, there is no option for a thee way portfolio TMF/UPRO/TQQQ. Like I said, I didn't really mess with the calculations, just added some automation and a few features for convenience. But I suspect it wouldn't be too hard to alter the calculations to account for a third asset. I might have a look later this week.

Btw, can anyone provide a quick explanation of the difference between minimum variance and inverse volatility? Why use one over the other?
I got a trial for a website that does the calculations and shows the difference with a monte carlo sim and volatility portfolio did a little better.
What tickers? What period? What website?
Rotationinvest.com upro tqqq tmf, used 1 month volatility and 6 months variance since inception of tickers.
User avatar
kerstverlichting
Posts: 27
Joined: Tue Feb 11, 2020 5:26 am
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by kerstverlichting »

Perfect Uncertainty wrote: Mon Sep 07, 2020 9:54 am
kerstverlichting wrote: Mon Sep 07, 2020 8:52 am btw, what's with the talk about using 6 mos worth of data
- Weight the 2 assets according to minimum variance optimization, using a “weighted” covariance matrix calculated based on 6 month correlation and 1 month (21 trading day) volatility.
It's not used in this calculation. Should it be added?
The calculations obtain the exact same AAA values as PV.

I use Excel on a Mac so maybe that’s an issue but the 3 queries you have error out. Maybe there is something you didn’t include in your version?
Hi, sorry, could you please clarify? How does this one month of data calculate 6 months correlation?

As for the error, it's probably the Mac version causing a fit. I intentionally avoided any vba and some of the newer functions (xlookup) for optimal compatibility but I need to import the yahoo data somehow... Maybe someone else can test on windows to confirm.
perfectuncertainty
Posts: 126
Joined: Sun Feb 04, 2018 7:44 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by perfectuncertainty »

The spreadsheet sample I provided uses 21 days of volatility.

If you want to change it you will need 127 days of data (127 for daily returns and 126 for Volatility and the Correlation coefficient).
Mickelous
Posts: 132
Joined: Mon Apr 20, 2020 11:24 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Mickelous »

perfectuncertainty wrote: Mon Sep 07, 2020 9:02 pm The spreadsheet sample I provided uses 21 days of volatility.

If you want to change it you will need 127 days of data (127 for daily returns and 126 for Volatility and the Correlation coefficient).
Which works just fine on the inverse volatility portfolio, which performs pretty well in the backtesting.
perfectuncertainty
Posts: 126
Joined: Sun Feb 04, 2018 7:44 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by perfectuncertainty »

I personally think that including 6 months of volatility to set a portfolio today isn't the best route. We are rebalancing monthly with AAA so I prefer the volatility calculation to be shorter in duration.
AnilG
Posts: 63
Joined: Wed Sep 09, 2015 1:41 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by AnilG »

I am using TQQQ/TMF strategy with monthly rebalancing using inverse volatility with 20 day volatility.
perfectuncertainty wrote: Mon Sep 07, 2020 10:56 pm I personally think that including 6 months of volatility to set a portfolio today isn't the best route. We are rebalancing monthly with AAA so I prefer the volatility calculation to be shorter in duration.
User avatar
kerstverlichting
Posts: 27
Joined: Tue Feb 11, 2020 5:26 am
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by kerstverlichting »

Alright, thanks for the explanation. I can only load 100 days of data anyway using the Yahoo connection. I will try to add the following things though:
- Option for 3-way TMF/UPRO/TQQQ, need to be able to set the UPRO/TQQQ split %
- Option for quarterly rebalancing, using 63 days of data to calculate volatility (still within the Yahoo limits), ability to easily switch between the two.
Last edited by kerstverlichting on Tue Sep 08, 2020 5:43 am, edited 1 time in total.
Post Reply