Jump to content
dydo

Trading Strategy

Recommended Posts

Event update. I will update more frequently now to keep track of whether TAN is clearly diverging from market.

 

return.png

Share this post


Link to post
Share on other sites

Man.. going full blown diversified is .. tedious. I'm analysing the properties of 250 stocks now after vetting maybe 1000. After that I have to optimize the allocation weights of these 250 stocks. The portfolio is getting a bit quanty.

 

Edited by explo

Share this post


Link to post
Share on other sites
On ‎6‎/‎12‎/‎2018 at 10:57 AM, explo said:

I'm analysing the properties of 250 stocks now

I'm starting to get my sheet together.

I think I need to buy a more powerful PC again.

Share this post


Link to post
Share on other sites
9 hours ago, explo said:

I'm starting to get my sheet together.

I think I need to buy a more powerful PC again.

Get your sheet together man!  You could use hadoop or the cloud?

Edited by sunnypease

Share this post


Link to post
Share on other sites
48 minutes ago, sunnypease said:

Get your sheet together man!  You could use hadoop or the cloud?

Yeah, the data is getting a bit big. Borderline that I need to manage it like that.

Share this post


Link to post
Share on other sites
On 6/7/2018 at 11:36 AM, explo said:

Event update. I will update more frequently now to keep track of whether TAN is clearly diverging from market.

 

return.png

So far the TAN is undecided. Without knowing anything about TA it looks like it’s at the cross roads to confirm a small double top peak before a crash or that it has entered a period of horisontal churn. Unfortunately solars are prone to fly and crash but some horisontal churning before can confirmation happen. Normally the churning does not precede a continued direction but a direction change. Cyclical style.

Note that TAN is broad solar, so inverter makers, manufacturing equipment makers and project developers are all included, not just panel makers. Manufacturing equipment makers usually take the biggest revenue hit earliest.

This is the solar index that TAN invests in: http://www.macsolarindex.com/

On yahoo the data is still better for the TAN etf than the ^SUNIDX index.

https://finance.yahoo.com/chart/TAN#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

 

Edited by explo

Share this post


Link to post
Share on other sites
On ‎6‎/‎12‎/‎2018 at 10:57 AM, explo said:

Man.. going full blown diversified is .. tedious. I'm analysing the properties of 250 stocks now after vetting maybe 1000. After that I have to optimize the allocation weights of these 250 stocks. The portfolio is getting a bit quanty.

 

I've gotten to the execution phase after a similar even more tedious process for funds. The asset allocation is now finalized and I bought over 100 different stocks on NYSE and Nasdaq today. There is no turning back from this. My stock basket is now finished. The funds are next - 70 of them have been allocated. Some of them will have delayed transactions.

Share this post


Link to post
Share on other sites
2 hours ago, explo said:

I've gotten to the execution phase after a similar even more tedious process for funds. The asset allocation is now finalized and I bought over 100 different stocks on NYSE and Nasdaq today. There is no turning back from this. My stock basket is now finished. The funds are next - 70 of them have been allocated. Some of them will have delayed transactions.

This sounds like an one-man EXPLO mutual fund.  How do you find time and energy to go through them all?  Are you starting taking clients?😀  Good luck!

Share this post


Link to post
Share on other sites
9 hours ago, Jetmoney said:

This sounds like an one-man EXPLO mutual fund.  How do you find time and energy to go through them all?  Are you starting taking clients?😀  Good luck!

Yes. It feels more like a one-man fund than a portfolio now with the very long list of assets. It has taken a lot of time and energy this year, but the idea is that after initial allocation the management should not be to much effort (it would be very effortless without the scale of assets). Even the initial effort only analyzed histoical assets returns, not their fundmentals. The analyzis is quite automated and the opimization based on the analysis is semi-automated once all the data to analyze has been prepared and all the analysis components designed and implemented. It will need many years to conclude if successful or not, but the  result data that I’ve gathered since start of 2016 (when I began balancing work) might change with this much higher degree of diversification.

Share this post


Link to post
Share on other sites
13 hours ago, Jetmoney said:

Are you starting taking clients?

I imagine that's a much larger (legal) threshold to climb. It sure seems like easy money to raise a billion investment for a fund and take 2% of it every year and 20% of its return above index. There must be some effort to it. I'll be happy with my own wealth growth if the strategy works. Then there's no need for a chunk of others.

Share this post


Link to post
Share on other sites
On ‎6‎/‎25‎/‎2018 at 9:24 AM, explo said:

So far the TAN is undecided. Without knowing anything about TA it looks like it’s at the cross roads to confirm a small double top peak before a crash or that it has entered a period of horisontal churn. Unfortunately solars are prone to fly and crash but some horisontal churning before can confirmation happen. Normally the churning does not precede a continued direction but a direction change. Cyclical style.

Note that TAN is broad solar, so inverter makers, manufacturing equipment makers and project developers are all included, not just panel makers. Manufacturing equipment makers usually take the biggest revenue hit earliest.

This is the solar index that TAN invests in: http://www.macsolarindex.com/

On yahoo the data is still better for the TAN etf than the ^SUNIDX index.

The TAN divergence became a bit more confirmed this week:

return.thumb.png.cca6e490c0935dd1ab57d398e3af4379.png

Edited by explo

Share this post


Link to post
Share on other sites
On ‎8‎/‎2‎/‎2018 at 11:45 PM, explo said:

I've gotten to the execution phase after a similar even more tedious process for funds. The asset allocation is now finalized and I bought over 100 different stocks on NYSE and Nasdaq today. There is no turning back from this. My stock basket is now finished. The funds are next - 70 of them have been allocated. Some of them will have delayed transactions.

Seeing clearly how massive diversification alters the portfolio properties I've restarted the portfolio with its inception being August 1 2018 (previous inception was January 1 2016). There are still some incomplete fund transactions. The major one is sale of a previously over allocated very volatile CTA fund. It will be sold next month. If it can behave until then (not influence the fund basket volatility too much) the inception point will hold in terms of correctly reflecting the start of the new massive diversification strategy.

I can't believe it took me 20 years of investing to learn what everyone was teaching - risk diversification is the free lunch of investing and it is stupid to not exploit it. Success yet to be proven..

return.thumb.png.e54f9417b9826a0dfd921173d4eaffc7.png

Edited by explo

Share this post


Link to post
Share on other sites

I'm changing my benchmark index from ^SP500TR to ^SP500NTR. While the inclusion of the dividends made the ^SP500TR a more fair benchmark than just the S&P 500 price index it was not fair to let the benchmark be free of the cost of tax (which the ^SP500NTR adjusts for).

The new massive diversification strategy (that started August 1 this year) has given the portfolio a real boner start chart (see below) and it's all alpha driven.

return.thumb.png.57ecb0562e882b6a2839c873f45c1e46.png

Edited by explo

Share this post


Link to post
Share on other sites

One month into my new massive diversification strategy I'm changing my objective to be much more conservative about my ability to generate future Alpha return. I think it is too easy to generate it in back-testing of a historical fact influenced strategy compared to do it in the test of time ahead. I'm therefore cutting my objective in half which extends the time-line for my goal with 5 years. The strategy and tactic remain unchanged. Here's the new GSOT:

 

Goal

Financial Independence 2030

Strategy

Explosive Growth Portfolio

Objective

10% Annual Excess Return

Tactic

20% Annual Risk

 

The new objective was almost hit already during the inception month which concluded yesterday:

 

return.thumb.png.c6ada0c9133593000dd73206eed47043.png

 

I'm moving away from the risk-parity concept, by considering the possibility that the two risk components of the portfolio - Alpha and Beta - might not offer equal reward. Accordingly, I've changed my profile picture to reflect this view.

I'm calling the new profile picture "Mr Risk". It has 4 curves. The two black legs at the bottom are the two risk components of a portfolio. The Alpha risk is the bent leg and the Beta risk is the straight leg. The black top of the head is the sum of them and the constant very light grey line between the connection points of the other 3 is the total risk. The picture thus shows portfolios of the same total risk but with different allocations of the two risk components.

Beta is the part of the portfolio risk that comes from its benchmark. My benchmark intends to reflect all available equity investments in the world - the most proven asset class for high long-term return. I've chosen to use a major US stock market index, the SP500NTR, as proxy for that. Alpha is the rest of the portfolio risk. It is defined by how the portfolio weights of assets from both the benchmark asset class and other asset classes differs from the asset weights of the benchmark.

Since Alpha and Beta risks are by definition zero correlated they are consequently far from perfectly correlated, which is what is required for the head of "Mr Risk" to not be completely flat. The lower the correlation is the higher the head is. This head is what is often referred to as "the free lunch of investing". 

Risk parity is where the legs cross and where the head is at maximum height. So parity, equally sized risks, gives maximum sum of risks for the same total risk. However, if the reward is not equal for the two risks this might not be the point of optimal risk allocation. I've chosen my allocation to be slightly to the left of that, which means that I am confident enough in my ability to generate Alpha that is more rewarding for the risk taken than Beta, but I'm humble enough to not be certain about it.

In numbers I'm using my total 20.00% risk allocation to allocate 16.00% Alpha risk which leaves 12.00% Beta risk. My risk sum is thus 28.00%, just slightly below the maximum of 28.28% occurring at parity allocation of 14.14% of each risk. This is the optimum when Beta is 25% less rewarding than Alpha, e.g. if 1% risk is rewarded with 0.4% Alpha return and 0.3% Beta return.

Above reward examples are the assumptions I've used for the new objective since 16% x 0.4 + 12% x 0.3 = 6.4% + 3.6% = 10%. In total the portfolio is assumed to reward 1% risk with 0.5% return since 20% x 0.5% = 10%.

Note that the risk-free return is added on-top of the 10% excess return objective. Historically this has averaged around 3.5% or 0.5% above inflation. With the new 2% inflation targeting monetary policies by central banks around the world the average risk-free return this century can be expected to be less than past century but at least be around 2% if the average inflation this century hits close to target. The long-term wealth growth rate target to beat is therefore closer to 12% than 10%.

The excess return objective of the portfolio is around 1.5 times higher than what the stock market has generated historically. The same applies to the reward per unit risk taken since my risk target is basically on par with the historical risk of the stock market. This is an aggressive market beat ambition but not too aggressive like the previous target of 3 times higher than stock market reward per unit risk taken.

I can still hope to achieve the old objective by being able to produce 1% return per 1% Alpha risk taken. Note that Alpha is the only risk where the reward per unit risk taken can be impacted by asset allocation. So, besides the comparatively easy task of the high-level allocation of Alpha risk vs Beta risk described above, this is where portfolio management can impact return by doing a good job on the low-level allocation of all the Alpha parts. The possible Alpha slices seem endless and the sum of them all has zero return and risk, so a really good but very possible job is required to allocate an Alpha risk that is more rewarding than the Beta risk. Defining the Alpha risk through asset allocation is basically all the work in portfolio optimization. Once it is done it is a simple decision to determine how much of it should be allocated versus the Beta risk.

For the non-zero-sum game of Beta only the amount of risk taken can be controlled unless you control all companies in the world. This still gives control of the amount of expected return from the Beta component. This is still a lot of control compared to the risk-free return component of your wealth growth which cannot be controlled in any way unless you are the head of a central bank without an inflation targeting policy that you must follow.

So although I can hope that I reach my old objective by finding an Alpha risk that is 3 times as rewarding as Beta (the stock market) I will not be optimizing my allocation of risk based on such hope. Reward optimum based on such hope would be a much more biased 19% Alpha + 6% Beta risk. Beta is actually much sweeter than its average most of the time, so even if the total outcome is slightly suboptimal due to too high Beta allocation the ride towards it might be more enjoyable more of the time. Maybe an underestimated trait in an allocation strategy. One that makes you stick to the strategy better. And more important than the strategy chosen is to stick to one strategy.

Allocation of too much Beta even if it is assumed to be more rewarding than Alpha would on the other hand make those few times when it is not sweet an extremely bitter experience for the portfolio. If there is a sufficient Alpha counter-balance in the portfolio this can be neutralized in order to avoid the pain and agony of deep and prolonged drawdowns. The right balance offers less risky and more consistent reward. Allocating Alpha can raise the return trend for a given risk level by the free lunches offered by low correlations, while allocating Beta can enforce trend fidelity by correcting inefficiencies in broad market risk pricing of returns. The first can be too uncertain and the latter can be too violent. So even in the case when the positive side of one of the two is preferred a balanced allocation might still be best in order to not be too exposed to of any of their negative sides.

Share this post


Link to post
Share on other sites
On ‎8‎/‎13‎/‎2018 at 6:54 PM, explo said:

Seeing clearly how massive diversification alters the portfolio properties I've restarted the portfolio with its inception being August 1 2018 (previous inception was January 1 2016). There are still some incomplete fund transactions. The major one is sale of a previously over allocated very volatile CTA fund. It will be sold next month. If it can behave until then (not influence the fund basket volatility too much) the inception point will hold in terms of correctly reflecting the start of the new massive diversification strategy.

Since the portfolio will not reach full diversification until above mentioned CTA fund gets sold this month the August 1 date is not very relevant as inception point. Instead I'm looking to incept this new strategy at the first point where the portfolio start behave according to it (meaning earliest time that concentrated holdings have been behaving since until sold off). I found this date to be March 23 this year, since prior that the concentrated holding in the CTA funds caused non-diversified behaviour of the fund basket.

Here's the new chart for new portfolio inception:

return.thumb.png.80dc5f322a48833d4fcbd22bbbdc4863.png

Edited by explo

Share this post


Link to post
Share on other sites
3 hours ago, explo said:

Since the portfolio will not reach full diversification until above mentioned CTA fund gets sold this month the August 1 date is not very relevant as inception point. Instead I'm looking to incept this new strategy at the first point where the portfolio start behave according to it (meaning earliest time that concentrated holdings have been behaving since until sold off). I found this date to be June 14 this year, since prior that the concentrated holding in the CTA funds caused non-diversified behaviour of the fund basket.

Here's the new chart for new portfolio inception:

return.thumb.png.a1950358bba8d73492e4778c633b4f81.png

Awesome! Very explosive growth, almost 20% in 3 months.

Share this post


Link to post
Share on other sites
5 hours ago, Jetmoney said:

Awesome! Very explosive growth, almost 20% in 3 months.

Thanks.  I hope 2018 can make up for the two lost prior years. I edited inception to March 26. That's when the CTA fund had cooled off after going haywire during the Q1 volatility.

Edited by explo

Share this post


Link to post
Share on other sites
On 8/17/2018 at 9:16 AM, explo said:

I’m changing my benchmark index from ^SP500TR to ^SP500NTR. While the inclusion of the dividends made the ^SP500TR a more fair benchmark than just the S&P 500 price index it was not fair to let the benchmark be free of the cost of tax (which the ^SP500NTR adjusts for).

Since the benchmark return must be in the same currency as the portfolio to be relevant and I’ve chosen to only include the stock market risk in it and not a currency risk I have used a hedged version of it, i.e. its return from value change in its nominal currency (USD) rather than in the portfolio currency. I was not considering the hedging cost before which primarily is the interest rate differential between the two currencies. I’m doing that now and call this benchmark ^SP500NTR (Hedged). 

Edited by explo

Share this post


Link to post
Share on other sites
On ‎8‎/‎13‎/‎2018 at 6:54 PM, explo said:

There are still some incomplete fund transactions. The major one is sale of a previously over allocated very volatile CTA fund. It will be sold next month.

It got sold today. My fund basket is finally diversified which makes my portfolio fully diversified now.

Share this post


Link to post
Share on other sites
On ‎9‎/‎1‎/‎2018 at 1:49 PM, explo said:

I'm changing my objective to be much more conservative about my ability to generate future Alpha return

I'm including the expected long-term risk-free return in my annual return objective and setting it to 15%. My allocation of those 15% are 3% from Risk-Free sources, 3% from Beta sources and 9% from Alpha sources. This allocation of expected return from different sources is now reflected in my percentages to the left. So far I'm ahead of both the benchmark and the target.

return.thumb.png.eb247cb1f61083283e0d7f6ef7068f4c.png

Edited by explo

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.



  • Donate

    Please donate to support this community. We appreciate all donations!

    Donate Sidebar by DevFuse
  • Upcoming Events

    No upcoming events found
  • Forum Statistics

    • Total Topics
      55
    • Total Posts
      90,738
  • Who's Online (See full list)

×