Jump to content
[[Template core/front/custom/_customHeader is throwing an error. This theme may be out of date. Run the support tool in the AdminCP to restore the default theme.]]

Muck's Model


muck
 Share

Recommended Posts

I've been putzing around with some new quantitative models here recently and have come across something that looks quite good ... but, because I've been offered a new job (still a little tentative about it), until I resolve whether or not to officially wind up my investment business to go do it, I'm not going to be trading the signal, so, I thought I'd pass it on to you free of charge in the interim.

 

Caveat emptor.

 

*****************************************************

 

Thesis: Markets are driven by fundamentals and psychology. Build a comprehensive model that takes both factors into account.

 

Specifics: Signal is derived from both fundamental data and technical data (psychology). Specific information analyzed (in a proprietary fashion) includes moving averages, PE ratios, dividend yield, 5yr US treasury yield, projected multi-year return on S&P500, and market sentiment.

 

Frequency: Trade Friday afternoon at the close. No mid-week trades. No stop-losses. No windfall profit rules.

 

Signal: Either you are 100% long the market or 100% short the market. No scaling in or out. No bonds. No international. S&P500 only.

 

Back-Tested Results: Since the summer of 1984, the signal results in an annualized return in excess of 21% / yr (some years are better and some are worse; some are negative). The signal made money in almost 3 of 5 weeks (59.4%). Best week was +18.2%. Worst week was -12.2%. The results do not account for transaction costs, slippage, margin interest paid on short positions,fees, etc. and they do not include dividends received (or paid if short).

 

Generally, the signal is long, however, there have been two periods where the signal was predominately short -- November 6, 2000 to July 26, 2002, the signal was short 75 of 90 weeks ... and ... May 19. 2008 to March 13, 2009, the signal was short 32 of 43 weeks.

 

I will try to this thread before the close of the market each Friday afternoon going forward for those that are interested... So, if you are interested, it'll be up to you guys to bump this thing because if it's not up top, I'll assume nobody is using the info and I won't waste my time. :wacko:

 

Recent signals:

SHORT 8/13/2010

SHORT 8/20/2010

LONG 8/27/2010

SHORT 9/3/2010

LONG 9/10/2010

LONG 9/17/2010

LONG 9/24/2010

LONG 10/1/2010

LONG 10/8/2010

Edited by muck
Link to comment
Share on other sites

  • Replies 300
  • Created
  • Last Reply

Top Posters In This Topic

What/which sentiment indicators are you using to determine psychology? What weight are you placing on fundamentals and psychology? What factors steered you toward a weekly timeline? Last but not least, why would you work for someone else when you have an investment company of your own? Seems like working for somebody else would be regressing at this point.

Link to comment
Share on other sites

What/which sentiment indicators are you using to determine psychology?

 

Not telling you any specifics beyond "I have some unique ways this model looks at moving averages and their derivatives and constituents". And, I'm not telling you what it is that goes into the moving averages (i.e., I am not disclosing what exactly it is that I'm averaging either).

 

What weight are you placing on fundamentals and psychology?

 

Approx. 50/50

 

What factors steered you toward a weekly timeline?

 

Nothing; I'm lazy. That and I don't want to be tied to the thing all day every day; I have other things to attend to. I just simply set up my in sample and out of sample data to look at things on a weekly basis...that's it.

 

Last but not least, why would you work for someone else when you have an investment company of your own? Seems like working for somebody else would be regressing at this point.

 

Seems like you also do not know of the opportunity I've been offered either. :wacko:

 

*******************************************************

 

In all seriousness, if I do not take this new opportunity, I may be launching a new investment fund with this model as the core component. For now, I'm offering it to the huddle tailgate for free if you want it ... all I ask is that if anyone wants to see the signal for the week to keep this thing bumped towards to the top of the tailgate on Friday afternoons and I will do what I can.

 

No promises I'll be able to get to it before the close, but I'll try.

 

No promises that it'll make you any money.

 

And I won't be offended if nobody wants to see it. :tup:

Link to comment
Share on other sites

  • 2 weeks later...
Not telling you any specifics beyond "I have some unique ways this model looks at moving averages and their derivatives and constituents". And, I'm not telling you what it is that goes into the moving averages (i.e., I am not disclosing what exactly it is that I'm averaging either).

 

 

 

Approx. 50/50

 

 

 

Nothing; I'm lazy. That and I don't want to be tied to the thing all day every day; I have other things to attend to. I just simply set up my in sample and out of sample data to look at things on a weekly basis...that's it.

 

 

 

Seems like you also do not know of the opportunity I've been offered either. :wacko:

 

*******************************************************

 

In all seriousness, if I do not take this new opportunity, I may be launching a new investment fund with this model as the core component. For now, I'm offering it to the huddle tailgate for free if you want it ... all I ask is that if anyone wants to see the signal for the week to keep this thing bumped towards to the top of the tailgate on Friday afternoons and I will do what I can.

 

No promises I'll be able to get to it before the close, but I'll try.

 

No promises that it'll make you any money.

 

And I won't be offended if nobody wants to see it. :tup:

Bump. I'm interested in what it projects for next week. I just put on a short trade into this rally.

Link to comment
Share on other sites

Does anyone here besides Muck have any clue what to do with this? Very generous of Muck to provide this but is it any use to us 401k slaves? :wacko:

 

I was thinking about that, and if I'm understanding everything correctly, you could have a chunk of your 401k set aside and on friday, if muck's thing said long you could throw it all in like a common stock fund, and if it said short you could throw it in like a fixed income or govt securities fund. just shift it back and forth accordingly. as long as they execute those requests in a timely manner (usually they trade at the closing price for that day) you could "play along" fairly effectively I think.

Link to comment
Share on other sites

I was thinking about that, and if I'm understanding everything correctly, you could have a chunk of your 401k set aside and on friday, if muck's thing said long you could throw it all in like a common stock fund, and if it said short you could throw it in like a fixed income or govt securities fund. just shift it back and forth accordingly. as long as they execute those requests in a timely manner (usually they trade at the closing price for that day) you could "play along" fairly effectively I think.

Some, if not all, of my funds have penalties for moves over a certain amount so I'd have to limit how many times I played this game. It'd then be like the lottery - the week you don't do it, up comes your number.

 

Might be better to have a little fund separate from the 401k and play via Schwab or something? :wacko:

Link to comment
Share on other sites

Some, if not all, of my funds have penalties for moves over a certain amount so I'd have to limit how many times I played this game. It'd then be like the lottery - the week you don't do it, up comes your number.

 

Might be better to have a little fund separate from the 401k and play via Schwab or something? :wacko:

 

as muck said, the signal is long most of the time. how many moves are you allowed over what time frame before the penalties kick in? you probably wouldn't usually use more than a couple a month.

Link to comment
Share on other sites

as muck said, the signal is long most of the time. how many moves are you allowed over what time frame before the penalties kick in? you probably wouldn't usually use more than a couple a month.

I don't know offhand, I'd have to look them all up and check. Good point about not having to do anything as long as it's the same signal from week to week.

Link to comment
Share on other sites

I don't know offhand, I'd have to look them all up and check. Good point about not having to do anything as long as it's the same signal from week to week.

 

When it does go short, it tends to trade long/short flip-flops quite a bit ... so maybe something like:

 

short

short

long

short

short

short

short

short

long

short

short

long

short

short

short

short

short

short

long

short

long

short

long

long

short

short

short

short

short

long

short

short

 

...etc...

 

...just so you can knowwhen you're looking into whatever it is that you're going to look into...

Link to comment
Share on other sites

What is it that you know about my model that would indicate that you think it'll be short next week?

My read on you is that you're good at forecasting, that's all. I think one of the next two Friday's your model will show a sell signal.

 

E2A: I think your updates should be new posts that are un-edited just to confirm time of post and give the thread a bump.

Edited by Brentastic
Link to comment
Share on other sites

My read on you is that you're good at forecasting, that's all. I think one of the next two Friday's your model will show a sell signal.

 

E2A: I think your updates should be new posts that are un-edited just to confirm time of post and give the thread a bump.

 

I ran the numbers last night; less than 1 in 6 weeks are "short" signals.

 

I'm not going to try to forecast the forecast, but I will tell you (and everyone else) that short signals tend to come in bunches and rarely (if ever) do they show up after getting 5-10 consecutive weekly buy signals after the most recent sell signals.

 

Because the model is a function of a variety of inputs (both fundamentals and price), it's possible that it could move to a sell in the next two weeks, but I would be a little surprised if it did. :wacko:

 

We'll see.

 

PS - Remember, this thing is only 'right' about 58% of the time, so, it'll be wrong plenty of times ... I guess the thing it's done best (historically) is tell you when it's a good idea to get out of the way (and go short, if that's your thing).

Link to comment
Share on other sites

Still long.

And as of right now, last weeks call of long is off (by only 1 S&P point). Granted, there is still some time for the market to rally a bit. Not calling you out in any way, just stating what I see. Looking forward to next week though, with elections and Fed meeting, it could get interesting.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...

Important Information