NoFeeBoards.com!

After numerous requests by others and considerable procrastination on my part I have recollected or regenerated most of the graphs that got lost from some of my earlier posts after the MSN site went belly up. I have (temporarily I hope) put some of my studies on a bare bones free web server while I look for the optimal permanent webhosting solution. Sorry about the obnoxious banner add at the top of each page. :roll:

Included is my enhanced Monte Carlo methodology post, portfolio rebalancing study, and an attempt a statistical proof that mean reversion exists.

The rebalancing study includes some information on the effect of diversification on SWRs which I know a lot of people are interested in. The mean reversion page is mostly new material and I would appreciate some feedback from those of you brave enough to wade through it, particularly those who are better mathematicians than I.

I hope these pages are of interest to some of you and I plan to add more soon. BTW, I'll be leaving town for 5 days tomorrow afternoon so I won't be able to respond to comments or questions after about noon tomorrow. I thought I'd go ahead and post the stuff anyway. :lol:

http://raddr.freehostingpro.com

excellent work, raddr! great to see those materials again.

enjoy your time away.

Hello raddr,

There is a banner ad, but no popups! I need to reread your "When to Rebalance?". I think I may still have to rebalance in Jan 2004. I guess I should just be happy that I may need to rebalance due to microcaps going up not down.

Oliver

PS Instead of rebalancing, I could just change my asset allocation. What is the max amount of microcap that can be justified? Would 41% of equity be considered a large % for microcap? Of course, one of the examples shows 100% of equity being ScV!

excellent site as it is, I am looking forward to more will add this to the FAQ work in progress

Oliver wrote: What is the max amount of microcap that can be justified? Would 41% of equity be considered a large % for microcap? Of course, one of the examples shows 100% of equity being ScV!

Hi Oliver,

I think every person has a different tolerance for this sort of thing. ScV is my largest asset class holding at about 30-35%. I own a lot of asset classes and have decided not to let any one class get much larger than about 1/3 of my portfolio just in case something bad happens to that sector of the market for an extended period of time. YMMV.

I just put up the "Importance of the sequence of returns for calculating SWR" study (http://nofeeboards.com/boards/viewtopic.php?t=391) on the new site if anyone is interested: http://raddr.freehostingpro.com/Sequence.htm.

Greetings raddr :)

I'd like to re-extend my offer to host your information here at NFB. You can have your own exclusive section right off of our home page. I hate to see your stuff in it's present form when I know we could do a super job presenting it here. Of course it's up to you. Let me know.

ElSupremo wrote: Greetings raddr :)

I'd like to re-extend my offer to host your information here at NFB. You can have your own exclusive section right off of our home page. I hate to see your stuff in it's present form when I know we could do a super job presenting it here. Of course it's up to you. Let me know.

Hi ES,

Thanks again for the gracious offer. One thing I was hoping for, though it is not necessarily vital, wherever I end up is that Frontpage extensions would be supported. Would this be possible here?

Thanks!

raddr wrote: I hope these pages are of interest to some of you and I plan to add more soon. BTW, I'll be leaving town for 5 days tomorrow afternoon so I won't be able to respond to comments or questions after about noon tomorrow.

Hi raddr,

Beautiful!

I liked the new mean reversion stuff.

Hello raddr!

Thank you for sharing your research, I have always enjoyed reading your messages and have found your work very educational. I will be re-reading all the "papers" you have shared at the web-site this week-end.

Hope your work receives the attention it deserves and helps many more make better financial decisions.

Regards,
Srini

Greetings raddr :)

I'll send you an email and we can discuss the details.

ElSupremo wrote: Greetings raddr :)

I'll send you an email and we can discuss the details.

ES,

I just emailed you back. I'd like to discuss this further when I get back from Tucson Monday nite.

Very informative stuff.

Raddr, thanks for sharing the info on your web site. I found it very informative. I have been running some of my own Monte Carlo withdrawal scenarios. What I found most interesting about your approach was the following:

From raddr web site:
What I did was create a random number generator where the standard deviation of returns was approximately equal to the historical norm. I set the average return to whatever number I want (more on this later). If the random number sequence is not within a certain tolerance I set then it is thrown out, either because the mean or standard deviation are off or there is no mean reversion tendency from the number sequence. This typically means rejecting around 97-99% of the number sequences.

The tendency to mean revert is manifest by the long term standard deviation (SD) of a return sequence being smaller than would be predicted from the short term SD. For example, if the one year SD is 18% then you would expect the 30 year SD to be 18/sqrt(30) or about 3.3%. In practice, however, it is about half that for most equity markets, including ours, likely because of mean reversion. To correct for inflation I use only real returns for the simulator.

I have struggled with this issue. Dr. Bernstein threw in the towel, so to speak, by just putting in a lower standard deviation number for his Monte Carlo analysis than the historical number, and citing return to the mean as the reason. I did the same. It turns out to make a big difference in the withdrawal number.

My thoughts about the whole thing are that I get very uncomfortable not assuming a normal distribution, but the evidence seems to suggest that the returns are not normally distributed (or is this because US was a "winner"?). It would be interesting to look at data from other countries -- this is how Fama justified the value premium in follow on studies to his seminal paper on the subject. I'll study your thoughts about mean reversion on the web site some more. Thanks for sharing.

My own guess is that (keep in mind I'm not a statistician) the distribution is leptokurtic, which means it has big tails. (Sorry for big word -- I did have to look it up). We have seen exchanges in Berlin, Paris, and Cairo go belly up this century, and major centers like Buenos Airies fall from grace. Looking at the past of a "winner", the US, means we have not seen the negative tail of the distribution, or maybe we have seen a positive tail. But if a negative tail comes, we are probably all screwed anyway. I guess it comes down to what Dr. Bernstein talked about, which is there are factors like catastrophic inflation, political failures, wars that mean there is no such thing approaching a 100% withdrawal rate, because the foundation of the system is not 100% safe, not even close.

Kramer

Hi Kramer,

Kramer wrote: the distribution is leptokurtic, which means it has big tails. (Sorry for big word -- I did have to look it up).

I use Gurunet to lookup words I am unsure about or plain don't know, right clicking the word. Free. Came up blank. Tried dictionary.com and it came up blank also!

peteyperson wrote: Hi Kramer,

Kramer wrote: the distribution is leptokurtic, which means it has big tails. (Sorry for big word -- I did have to look it up).

I use Gurunet to lookup words I am unsure about or plain don't know, right clicking the word. Free. Came up blank. Tried dictionary.com and it came up blank also!

Try this link, although if I'm reading it correctly, platykurtic connotes fatter than normal tails, leptokurtic the opposite.

I found a few URLs where the opposite claim is made; not sure why the contradiction--maybe a narrower (x-axis) bulge/peak in the frequency distribution graph leaves more (x-axis) room for the tails, and using loose terminology we could claim they are "bigger" tails. I believe leptokurtic is intended to refer to tighter clustering about the mean, platykurtic to a more uniform frequency distribution.

PITA

PainInTheAS wrote:Try this link, although if I'm reading it correctly, platykurtic connotes fatter than normal tails, leptokurtic the opposite.

In the quote above, the bolded word should be flatter.

Made the rest of the post rather incoherent...

PITA

PainInTheAS wrote:
In the quote above, the bolded word should be flatter.

Made the rest of the post rather incoherent...

I was wondering about that, because I thought I understood what you were talking about, then that bit threw me for a loop.

Guys & Gals,

Thanks for feedback and the kind words. I'm glad some of you are finding topics of interest on my site. I plan to keep adding topics. A study on the predictive power of the Gordon Equation (expanded from an earlier NFB post) is probably going to be my next project.

ES has offered to host the site here at NFB which sounds great to me. I plan to get with him soon to work out the details. When the site changes over I will of course include an updated link.

Greetings raddr :)

When the site changes over I will of course include an updated link.

LOL! :lol:I don't think they'll have any trouble finding it!