Bob Brinker Free Discussion Site 59,820+

Top 50. Feb 28, 2005 11:36 AM

» JIMMY62 - Re: Brinker, Off and On

In response to Brinker, Off and On posted by honeyoneohone:

Does this mean that the market has been going up, but at some point will go down? LOL!

Yes, also it is Bob's same old stuff.

By that I mean to say that Bob always says a market turn is likely within the time period of a one year subscription to his MARKETIMER newsletter.

-- posted by JIMMY62

Top 51. Feb 28, 2005 11:43 AM

» JIMMY62 - Re: Re: Brinker, Off and On

In response to Re: Brinker, Off and On posted by arommel88:

I think he does not expect to see the SP 500 to hit its 2000 high for some time. He is probably right.

IMO it is a widely accepted understanding that the 2000 high of the SP 500 occurrd at the peak of a speculative bubble.
Because of that it has is no significant value as a market indicator. Reaching it or not reaching it has no value of a predictor of appropriate market levels.

-- posted by JIMMY62

Top 52. Feb 28, 2005 12:26 PM

» bob90245 - Re: Re: 50:50 Equities:Bonds is UP 18% since 1/1/00

In response to Re: 50:50 Equities:Bonds is UP 18% since 1/1/00 posted by arommel88:

If anything, using bonds and balancing to smooth out stocks for people who cannot stomach the ups in downs should be very instructive.

Actually, I think the "secret sauce" is in the rebalancing. IOW, selling some from bonds in order to buy more shares of stocks when low will propel the portfolio further during the eventual recovery.

This is in contrast to just letting 50% stocks sit side by side with 50% bonds without rebalancing. I would strongly suspect that this latter strategy would not produce as good results as the former.

-- posted by bob90245

Top 53. Feb 28, 2005 12:47 PM

» arommel88 - Re: Re: Re: Brinker, Off and On

In response to Re: Re: Brinker, Off and On posted by JIMMY62:

I am not sure what that means? Markets have to endure speculation and it does not matter what the cause was to the reality of an investment. It is really not an indicator but a statement of opinion. If the bubble can be defined at the time of the event then you are a market timer and is not really a good basis to start riding BB. BB has claimed that we ented a bear market from the 2000 top. As I said, it is probably correct that we will not see that high this decade. One 18 month recession will run out the clock.

-- posted by arommel88

Top 54. Feb 28, 2005 1:16 PM

» arommel88 - Re: Re: Re: 50:50 Equities:Bonds is UP 18% since 1/1/00

In response to Re: Re: 50:50 Equities:Bonds is UP 18% since 1/1/00 posted by bob90245:

I was not suggesting that one not balance, it would not be anything new. The target allocation is where I suggest someone pick one so that they do not panic out of it. I think the "secret sauce" is really buying things that are out of favor that really do not impact intrinsic value. In statistics, it is called regression. Rebalancing is riding the wave of regression. It happens in many areas of our life. Typically children of tall parents are a little shorter etc.In reseach methods, when I did not find a correlation, I was always taught to pick the mean.

Kirk's plan seems like pretty good advice to ride this wave. If you have a talent like Warren Buffet then it seems that one can ride this wave more actively.

I think the bond market reflects an asset class that falls in and out of favor out of step with stocks and that is what makes using them in this way effective.

-- posted by arommel88

Top 55. Feb 28, 2005 2:19 PM

» Normxxx - Re: Re: Re: Re: 50:50 Equities:Bonds is UP 18% since 1/1/00

In response to Re: Re: Re: 50:50 Equities:Bonds is UP 18% since 1/1/00 posted by arommel88:

In statistics, it is called regression.

Yes, but that's only true of closed sets, and the 'future' market is an open and unbounded(?) set. In a closed set, we have regression (of the sample mean) to the 'true' mean (of the set)because the 'true' mean is an unknown, but fixed value. In an open, unbounded set, the 'true' mean changes as we 'add' each days sample to the set. 'Regression to the (existing) mean' may never occur-- the (existing) mean may merely rise or fall to meet the 'new' values.

Nevertheless, many measures of the market act like they are 'regressing to the mean' because outlying numbers are likely to be improbable (i.e., 'rare'). Just don't bet the farm on it. After coming up heads 100 times in a row, the probability of a 'true' coin coming up heads or tails on the next throw is still 50:50 (this is an example of an open, but bounded set).

-- posted by Normxxx

Top 56. Feb 28, 2005 3:24 PM

» arommel88 - Re: Re: Re: Re: Re: 50:50 Equities:Bonds is UP 18% since 1/1/00

In response to Re: Re: Re: Re: 50:50 Equities:Bonds is UP 18% since 1/1/00 posted by Normxxx:

market is an open and unbounded(?)

I would not say that. It is certainly poorly defined but we can theoretically max out inputs. However even with a main effect operating on the system you will still put up with regression, even if the "real" mean moves. You may very well fail to observe regression but it is there.

Nevertheless, many measures of the market act like they are 'regressing to the mean' because outlying numbers are likely to be improbable (i.e., 'rare').

You are talking about the probablity of a type I/II errors(false positive/negative errors) in statistics? Not really directly related to regression. Using regression is the opposite of trying to interprete data within your tolerences. In that case you are trying to interprete data instead of trying to benefit from the natural flow of regression. What you are referring to is a confidence level of an observed correlation and that fits a timing model. Bob Brinker looks for correlations with economic data and markets. Kirk suggests rebalancing which is a model that uses regression.

the probability of a 'true' coin coming up heads or tails on the next throw is still 50:50 (this is an example of an open, but bounded set).

Hence, the rebalancing every year. Don't bet on a single flip of the coin. Regression does not work with a single flip of the coin. What you are betting on is that you will pick up on something that should not be.

A hundred people taking a test is bound to have someone score higher because he decided to study that time. He will probably go back to his old ways. A low score may mean his girl friend who is normally out of town, spent the week end. She will not be there next time etc.

-- posted by arommel88

Top 57. Feb 28, 2005 6:13 PM

» bob90245 - Re: Caller: FICO Credit Score

In response to Caller: FICO Credit Score posted by Kirk:

Liz Pulliam Weston, the author of that article, has a Sunday column at the LA Times. I've been reading her column for many years. So I knew of these FICO tidbits. My FICO score is in the top 10% which I'm told is fairly rare. When I first moved into my apartment a few years ago, my roommate told me that the landlord had never seen a FICO score so high.

-- posted by bob90245

Top 58. Feb 28, 2005 8:56 PM

» pete2214 - Re: Re: Brinker, Off and On

In response to Re: Brinker, Off and On posted by JIMMY62:

double babble talk. Easy way to cover his behind.

-- posted by pete2214