Search | User Listing Vidsite | Calendars | Quotes
Home Page ->  Markets & Methods -> Mood of the Market -> View ThreadLogon (or Register, or Join TradeTight)

You are logged in as a limited-access Guest.To join TradeTight, first read the info in the Organization & Content room, then click the link above. 

Psalm 127:1 (NKJV) ... Unless the Lord builds the house, they labor in vain who build it; Unless the Lord guards the city, the watchman stays awake in vain.


Formal Evaluation of Market State
Jump to page : 1
Now viewing page 1 [50 msgs/pg]
Jump to room :
JimDean
Posted 10/27/2013 4:00 PM (#5299)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
For a long time I've been developing and refining an algorithm that evaluates the Trend, Volatility and Market State (combo of T&V), for general use in Filters, and possibly even longterm position-trading. It also will help with OVest, eventually.

It studies a weighted mix of the TVM factors for Market, Sector, Industry, Group and Symbol. The S,I,G info comes from the MG averages associated with the current Symbol being studied. But I'm sort of undecided about what symbol/index to use for the "Market" portion ... which is important since it's likely to receive considerable weight when combining the five components.

Three important requirements are that the Market symbol/index be available back into 1999 or earlier, and the data should be available with a standard OData EOD feed (most of the many "Indexes" listed in OScan are not, btw), and the symbol should be very liquid (or for an index, very widely examined).

Another factor which I'm of two minds about is whether to use a tradeable symbol such as SPY or QQQ versus the related index $SPX or $COMP. The difference is relatively small but can affect the results ... the bars (particularly Daily) of the tradeable instrument tend to show more gaps and doji's than the indexes, since the indexes are calculated derivatives.

So, I'd appreciate your feedback regarding which symbol(s) to use as the basis for the "Overall Market" viewpoint. I'm considering:
SPY or $SPX (SP-500)
IWM (Russell 2000 composite)
QQQ or $COMP (Nasdaq composite)

My current plan is to use an average of the results for SPY, IWM and QQQ as the Overall-Market component.

See the next post for some snapshots to consider.
Top of the page Bottom of the page
JimDean
Posted 10/27/2013 4:02 PM (#5300 - in reply to #5299)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
To help you evaluate this, I am attaching two snapshots based on WEEKLY bars ... the thread-display shows squished versions just as an overview ... please download the Zip file which has big versions that you can study more effectively.

The Zip file also has three sets of charts for each comparison ... based on Daily, Weekly and Monthly bars. For normal Daily-bar Swing trading, the Weekly Market State would be the Filter of choice. For intraday trading, the Daily Market State would be the right one to use. For longterm Position trading (hold time of a few months) - typically LONG-ONLY, the Monthly Market State would be the Filter of choice (note that due to lag, and since downtrends are usually short-duration, this does not "recognize" Bearish conditions until they are fairly mature).

SO ... for this purpose, please focus on the WEEKLY charts. They offer plenty of detail and yet not too much to digest.

The Indicator at the bottom has three main components. At the top there is an upside-down histogram showing Trend where Green=Bull, Red=Bear, Orange=Neither/Flat. At the bottom there is another histogram showing Volatility where Cyan=Wild, Navy=Calm, MedBlue=Typical. In the middle there is a plot of market state that should be ignored since those snapshots are actually incorrect.

Please pay attention to the TOP HISTOGRAM (green/red/orange) regarding the TREND.

There are charts shown for SPY vs SPX, as well as QQQ vs SPY vs IWM ... the main thing you should notice is that the Trend histogram varies significantly between them, using the same inputs for evaluation.

OK ... that should give you plenty to think about

Comments and suggestions are most welcome!


(SPXvsSPY Weekly.png)



(QQQvsSPYvsIWM Weekly.png)



Attachments
Attachments Six Fullsize Charts.zip (536KB - 28 downloads)
Attachments SPXvsSPY Weekly.png (83KB - 7 downloads)
Attachments QQQvsSPYvsIWM Weekly.png (103KB - 9 downloads)
Top of the page Bottom of the page
SteveLuerman
Posted 10/27/2013 5:58 PM (#5303 - in reply to #5299)
Subject: Formal Evaluation of Market State



Elite

Posts: 414
100100100100
Location:
USA: CO, Boulder
Jim,
A quick off-the-cuff comment (I haven't given your questions a lot of thought yet):

I'm sure you've considered, but here goes... Assuming this "market state filter" can be tweaked with parameters, you could make all these choices user-selectable with a parameter. One index or another may be more appropriate for one purpose vs. another.
Top of the page Bottom of the page
JimDean
Posted 10/27/2013 6:01 PM (#5304 - in reply to #5303)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Hi, Steve:

You're right in theory, but my plans for this tool preclude that solution ... this will be potentially used by many people who would not understand the significance of the choice, nor be familiar enough with the market in general to make a wise choice.

It will be possible for the person to use the tool to focus on just one symbol, which could be an index or tradeable instrument of one ... but the considerations I am dealing with now relate to cases where a weighted combo of OverallMkt, Sector, Industry, Group and Symbol is the desired goal.
Top of the page Bottom of the page
JohnB
Posted 10/28/2013 5:56 AM (#5305 - in reply to #5299)
Subject: Formal Evaluation of Market State


Friend

Posts: 29
25
Location:
: ,
Jim I do know from an ex Russell employee that the Russell 2000 index ($RUT) is considered to be the most representative market index and is used as such by the City of London dealers. Best Regards John B
Top of the page Bottom of the page
GregWinch
Posted 10/28/2013 10:45 AM (#5307 - in reply to #5299)
Subject: Formal Evaluation of Market State



Regular

Posts: 78
252525
Location:
USA: AZ, Phoenix
Jim,
A few comments from the bleachers:
1. I firmly believe using this type of analysis as a FILTER is the "missing ingredient" to success of most OT systems and strategies.

2. In the Swope / Howell book "Trading by Numbers", the authors used an average of $DJI, $SPX, $NDX, $MID & $RUT for their Trend Scoring system. Since $MID & $RUT are no longer provided by OmniData, the ETF surrogates of DIA, SPY, QQQ, MDY & IWM may be useful. Since you've used SPY, QQQ and IWM, your research may have shown that adding DIA and MDY might be overkill when averaging.

3. From your post here and previous posts of similar nature, the influence of the work of Van Tharp is evident. I believe that if we can conquer the 9 (or 12) market types as a filter, we will have a useful method in a backtesting environment to help determine what works best when.

As usual, great work to date on a monumental task!

Greg
Top of the page Bottom of the page
JimDean
Posted 10/28/2013 12:06 PM (#5308 - in reply to #5305)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Thanks for feedback guys!

I will reconsider DJIA - it doesn't cover many stocks or sectors but it has a big psych component. I will also consider including MDY - possibly the total weight of it and SPY would be even with the weight of IWM since they cover the same ground. Since QQQ is a very different kind of mix that represents a lot of stocks that are actively traded, it seems important.

I will also consider offering a few choices per Steve's suggestion. With five symbols there are 120 permutations so I'll probably prethink it a bit to keep the input simpler. Here are options:
0 = avg of all five, evenly
1-5 = each of the five, individually
6 = ((spy+mdy)/2 + (qqq+djia)/2 + iwm)/3
7 = ((spy+mdy)/2 + iwm)/2
8 = (spy+mdy)/2
9 = (iwm+ qqq + djia)/3

… there's (somewhat obscure) logic behind each of those … I am most drawn to #6 & #9, with #7 also pretty good … I also consider #8 sort of redundant (vs #7 or iwm by itself), #0 pretty arbitrary, and #1-5 sort of weak.

Any other suggestions?

Or other symbols to consider? I want I keep them domestic since OData doesn't currently provide coverage overseas. More symbols do somewhat slow down the process btw but not too much.
Top of the page Bottom of the page
JeffMcClure
Posted 10/28/2013 1:39 PM (#5310 - in reply to #5299)
Subject: Formal Evaluation of Market State



Friend

Posts: 28
25
Location:
USA: KS, Overland Park
Perhaps a couple of other elements to consider in the Market State recipe:

1. Inter-market relationships/confirmations - USD/Euro (etf UUP/FXE), Bonds (etf TLT/TBT), Volatility (etf VXX), Metals (etf GDX/SLW)
2. Relative strength of symbol/market ETF with respect to Market State/Index metric - i.e. if IWM is stronger than SPY/QQQ/Market State and market is breaking long may want to weight shares to IWM
3. Correlation of symbol/market ETF with respect to Market State/Index metric
4. Seasonality of symbol/market ETFs/Market State

I've been have recently been studying with some folks that take seasonality to another level (compared to Nirvana) and compliment it with channel trading. Could see something like what they do layering in quite nicely to Market State.

http://www.magentatrader.com/home/

Here is a free video where they review stocks on Sunday evening for the upcoming week in the context of seasonality. Interesting stuff.

http://www.magentatrader.com/webinars/free/2013-10-20%2021.00%20Sun...
Top of the page Bottom of the page
JimDean
Posted 10/28/2013 2:24 PM (#5311 - in reply to #5310)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Thanks Jeff. All are excellent ideas, geared mainly towards prospecting and portfolio management, &/or position sizing biases.

What I'm working on, otoh, is a very broadbased conceptual solution to the age old question "what is the market doing?" And my core premise is that the most useful information, with the least lag and noise, is derived directly from OHLC values, rather than from other things such as correlation, or advance/decline, or cycle theory, etc. Nothing's wrong with that other stuff ... but to be truly "universal", I'm focusing on OHLC-based analysis.

I have chosen to answer that question in two categories: Trend and Volatility, each of which have three conceptual states (with continuous inbetween scores as well). The three-state (Bull/Bear/Flat and Wild/Calm/Norm) thing diverges from classic dualistic market thinking, and that's really important imho. You can see the Trend as the upper histo and the Volat as the lower histo on the charts.

Using those two (T&V) components, I derive a third overall "Market" metric that has nine nominal states (Bull Calm, Bear Norm, Flat Wild, etc), again with continuous inbetween scores. You can see this as the middle plot in the charts I provided (even tho those charts are not completely correct since I've done a lot of work on them since).

I have already built that tool ... the thing that determines the TVM values ... I've working on it off and on for many years. Maybe a few small refinements and tweaks might be added but it's 98% locked in place. The tool can be applied to whatever symbol (with OHLC values) that seems useful. What I'm doing now is creating a shell routine that calls the core TVM calculator and weights the results from various representative "levels" of the trading environment, from the specific symbol on up through Group and Industry and Sector into the full market. I believe that price action of a symbol is USUALLY as much or more a function of some combo of those "higher" levels as it is an individual entity. So, the shell program allows for very flexible weighting of each of the levels.

The MG##'s provide very convenient and widely available measures of Group, Industry and Sector, with thorough mappings of symbols to the various levels. There is a fluctuating group of various ETF's that sort of does the same thing but since the makeup and availability of those various ETF's (and the way they are computed) is always changing, I've chosen to disregard them for the time being ... they could be added later though if a stable enough and universal enough structure evolves, along with built-in "mapping" from symbol-to-composite and vice versa (not currently avail in OT as thoroughly as necessary).

The MAIN purpose here is to create a FILTER to be used with most any trading system, and to create a BIAS FACTOR to be used with money management (portfolio balance and position sizing).

An "offshoot" of it is that the various metrics (TVM) can be used to create entry &/or exit signals ... but that is NOT my primary goal, so I'm not structuring things around that ... signals are derivative benefits, if they happen to work out.

I hope this explanation helps everyone to get a better feel for what I'm trying to do, so you can focus your comments on the specific questions I've asked, within that structure.

Thanks.
Top of the page Bottom of the page
SteveLuerman
Posted 10/29/2013 4:41 AM (#5314 - in reply to #5308)
Subject: Formal Evaluation of Market State



Elite

Posts: 414
100100100100
Location:
USA: CO, Boulder
Jim,
These indices range from approximately 80 (QQQ) to approximately 15,000 ($DJI). So, I would think you'd need to normalize them before combining them as described in combos 6 thru 9. You probably were planning to do that, but I thought it worth mentioning (just in case).
Top of the page Bottom of the page
JimDean
Posted 10/29/2013 8:08 AM (#5315 - in reply to #5314)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
The TVM analysis is run on each symbol individually. The results are normalized for each (T&V=10-30, M=10-90). Then the TVM values are combined as I described earlier. The pseudo-formulae for weighted mixes shown above are combos of TVM values, not of the raw OHLC values. Think of the TVM values as "scores". They most definitely are normalized.

Does anyone have comments on the combo's I suggested above? Or any further suggestions about core symbols/indices to represent the market?
Top of the page Bottom of the page
MelDickover
Posted 10/29/2013 8:38 AM (#5317 - in reply to #5299)
Subject: Formal Evaluation of Market State



Veteran

Posts: 155
1002525
Location:
USA: FL, Bradenton
To my way of thinking, one needs to know if the volume has been accumulation or distribution, and what the advance/decline line is doing.

Mel
Top of the page Bottom of the page
JimDean
Posted 10/29/2013 8:58 AM (#5318 - in reply to #5317)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Several have mentioned A/D and I do understand the reason. I had originally disregarded it since it is not a "true symbol". But when one thinks about it as a companion metric to volume, such that both can serve to reinforce or attenuate the direction/speed of consolidated price movement, that may improve the quality of the analysis.

I will give some thot to how those factors might be combined with the existing core algorithm. Basic and consistent concepts, not complex or flighty ones, are what I am rooting the analysis in. There are three such concepts that I can think of readily regarding the impact of A/D and Volume:

1. Rising (smoothed) A/D is bullish, falling is Bearish, and particularly choppy accentuates price volatility concerns.

2. Major isolated Spikes in Volume (ie 5-10x normal) need to be capped for averaging purposes, to see what is really going on behind the scenes.

3. Rising &/or High volume accentuates/reinforces price trend moves when in direction of the trend; sudden volume increases when price is contra-trend indicate transition to flat/ retrace/ consolidation that probably will last a few bars at least

If y'all have other similar "general" rules you'd like to share please feel free. I am not interested in "microscope" patterns that might be used as "signals" but rather for broader effects that speak to the State of the market as discussed earlier. Thanks.
Top of the page Bottom of the page
GregWinch
Posted 10/29/2013 10:12 AM (#5319 - in reply to #5299)
Subject: Formal Evaluation of Market State



Regular

Posts: 78
252525
Location:
USA: AZ, Phoenix
Regarding use of volume:
If ETF symbols are used in lieu of the actual indexes, one may consider ROC of On Balance Volume as a factor that can be "scored". Comparing the 20 day EMA of OBV to the 20 day EMA of OBV say 10 days prior for each ETF may be useful.

Regarding which combination of Symbols:
Is there a way to show correlation of the different combos to what the market is actually doing? I suppose the "correct" combo would be an individual choice depending on the user's trading tendencies (i.e. "I only trade large caps" would mean SPY or DIA deserve more weight). My vote is that for a universal type indicator a normalized average of all five would work best.

FWIW,
Greg

Edited by GregWinch 10/29/2013 10:18 AM
Top of the page Bottom of the page
JimDean
Posted 10/29/2013 10:48 AM (#5320 - in reply to #5319)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Thanks, Gregg. Yes, OBV tells the kind of story I'm looking for ... but I lose the ability to determine if a given bar's volume (or MA of it) is relatively "high" or "low" vs some long historical band or MA. Also, I'd lose the ability to "cap" the huge spikes unless I recalc'd OBV manually. So, I'll probably stick with raw V, using responsive MA's of it after capping. But thanks for the suggestion!

I'm not sure what you mean by "correlation" ... if you are speaking of statistical analysis then it's sort of a circular question ... the goal here is to DETERMINE the basis against which OTHER things might be correlated. Re the various Market measures ... if they are NOT correlated with one another well, then that is a reason to include them since by doing so it gives a better-balanced big-picture. HOWEVER ... if those individual indexes are not well correlated that could indicate greater indecision, which would attenuate trend strength and boost volatility scores.

Unfortunately correlation of multiple dataseries with one another is a pretty complex mathematical problem to solve. Individual pairs aren't so bad though. I suppose that I could compare each index to the weighted average of them, then average those correlations to get an overall metric. That's a lot of thrashing around code-wise ... not sure if benefit would be worth it. Will think more about it.
Top of the page Bottom of the page
JimDean
Posted 10/29/2013 11:13 AM (#5321 - in reply to #5320)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Here is the kind of thing I'm looking for suggestions about, re good representative overall-market candidates ... these three lists of course have too many choices ... I'd like your comments on them.

It's important that:
1. they should cover a wide base of related symbols (too avoid too much bias),
2. they should go back a LONG time in history, ideally to the beginning of 2000, for reliable historical studies.

See these links as a starting point:
http://money.usnews.com/funds/etfs/rankings/small-cap-funds
http://money.usnews.com/funds/etfs/rankings/mid-cap-funds
http://money.usnews.com/funds/etfs/rankings/large-cap-funds

Further, re the "weighted mixes" ... mention was made of people who "prefer" to trade small, mid or large caps. I don't generally use that as a trading filter, but I'm curious if a lot of you DO (for active trading rather than investing). If so, there could be pre-set mix options that focus on those categories.
Top of the page Bottom of the page
GregWinch
Posted 10/30/2013 11:21 AM (#5322 - in reply to #5320)
Subject: Formal Evaluation of Market State



Regular

Posts: 78
252525
Location:
USA: AZ, Phoenix
Jim,
I will try to explain what I mean by correlation. Assuming that what you are considering to be Market State has at least partially something to do with market trends (not considering Volatility for the moment) and looking at your red-green-orange indicator in the above post, the indicator should be orange (neutral) at market turns and green (bull) or red (bear) in between. Whichever combination of symbols (e.g. SPY+QQQ+IWM/3) used in the calculation best correspond to the actual changes in Market direction would "correlate" best.
Thinking in terms of major moves (long term trading) and looking at SPY for the "market", the indicator should be orange around March 2000, Sept 2002, Oct 2007 & March 2009 with red and green between these dates. The best correlator would be the one that is orange the least amount of periods at tops and bottoms, joining the direction of the new trend as soon as possible after market turns. This thinking, of course, can be applied to any time frame.
Market volatility can be added to the mix to further segment the market. Bull Normal, Bull Quiet, Bear Volatile,
Neutral Quiet, etc.
Am I on the same page as you or thinking down a totally different avenue?
Greg
Top of the page Bottom of the page
JimDean
Posted 10/30/2013 12:22 PM (#5323 - in reply to #5322)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Hi Greg - yes we are on the same page but possibly our glasses are tinted differently ;-)

Correlation compares similarly-granular dataseries. You are talking about "co-incidence" of two different kinds of things - price MA's vs pivot points (to oversimplify). MA's create natural lag to obtain smoothness. Pivots use a crystal ball.

If the MA of a particular combo of ETF/index symbols happens to coincide occasionally with pivots of that combo (ie the flip from green/red to orange) then that is serendipitous but not IMHO reliably predictive. Plus there is a chicken/egg thing - you're comparing the combo to itself to choose which combo is best.

What I'm asking for here is which combo of ETF/index symbols is MOST REPRESENTATIVE of the market as a whole, rather than which one(s) might have "led" the market direction at some time(s) in the past.

The way to analyze what you are speaking about would be a group-trader kinda thing. But GT does not have the particular option (yet?) that is needed. It should determine how many bars back/fwd a given symbol's dataseries should be shifted, to best correlate with other symbols in the group - thus pointing out leaders and laggards rather than just middle-of-road most-correlated.

Thanks for your comments but I keep on coming back to my original questions:
1. What ETF symbols are most reliably representative?
2. What weighting of them is most representative?

Thanks.
Top of the page Bottom of the page
AzizAsulaibi
Posted 10/30/2013 5:08 PM (#5324 - in reply to #5299)
Subject: Formal Evaluation of Market State


Friend

Posts: 17
0
Location:
Saudi Arabia: , Riyadh
Hi Mr.Jim and All

I don't trade in US market but I see this page ... i hope it help or make a new ideas

http://www.marketvolume.com/technicalanalysis/new_highs_lows.asp

best regards
Top of the page Bottom of the page
SalimHira
Posted 10/30/2013 7:12 PM (#5325 - in reply to #5299)
Subject: Formal Evaluation of Market State



Veteran

Posts: 183
100252525
Location:
USA: MD, Columbia
Hi Jim:

Fwiw, the markets are now Global - where monies are flowing in / out of U.S-based markets in matter of seconds. As such, you also have six major ETF currencies, Bonds, and above all, the US Dollar ETF (as the reserve global currency) which must also be taken into account to give a full scope of where the "wind" may be flowing inward or outwards from the U.S. based market to get a very good handle and perspective of what you may be attempting to accomplish.

This has nothing to do with correlations, etc. and also not sure, how it would be incorporated in your code, but thought, something to ponder upon as too no stone was unturned.

It is a very complicated challenging dilemma you have taken on :-), imho. Thank You.
Top of the page Bottom of the page
DonSchuman
Posted 1/12/2014 10:51 AM (#5547 - in reply to #5299)
Subject: Formal Evaluation of Market State



Veteran

Posts: 212
100100
Location:
USA: TX, Mansfield
This is a very interesting topic and important project.

Usually the most oft asked questions are the most difficult to answer.

I don't know whether or not this is in the ball park, but I haven't heard any mention of equal weighted indices.

Do they not fit into your scheme or vision for this filter?

I would love to know where you are with this filter and what your plans are for it.

Excellent work Jim!

Don
Top of the page Bottom of the page
JimDean
Posted 1/12/2014 2:19 PM (#5548 - in reply to #5547)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
Hi, Don

Status: I've got the hard legwork done and am working on how best to use it and to provide it to others. I'll get there.

Weighted indices: I'm definitely planning on providing "user options" for which indices to use, and how much weight each would be given. There are many ways of looking at it ... I want to make it simple but also very flexible.

I hope to have an initial release of this, in some form, later this Spring. As with several other things, this is waiting on the completion of the Vidsite work. The new Vidsite will provide for a LOT of additional capabilities.
Top of the page Bottom of the page
DonSchuman
Posted 1/12/2014 2:31 PM (#5549 - in reply to #5299)
Subject: Formal Evaluation of Market State



Veteran

Posts: 212
100100
Location:
USA: TX, Mansfield
Thanks for the update Jim.

I know you stay very busy.

Let me clarify my question about "equal-weighted indexes."

I see the discussion about SPY vs. $SPX and their counterparts. These Indexes and their derivatives are weighted (ie. appl accounts for a fairly large percentage of the $NDX).

What I'm wondering is whether or not you have considered using the equal-weighted indexes instead?

So instead of the $SPX, you would use the RSP (Guggenenheim S&P 500 Equal Weight ETF), etc.
Top of the page Bottom of the page
JimDean
Posted 1/12/2014 2:36 PM (#5550 - in reply to #5549)
Subject: Formal Evaluation of Market State



Owner/Admin

Posts: 3925
2000100050010010010010025
Location:
USA: GA, Lawrenceville
I'm certainly willing to consider any broadly-useful index that y'all might suggest.

You can save me some effort by providing a list of "equivalent pairs" (like Don just did, weighted vs unweighted) so that I don't have to look them up.

One requirement, please ... make sure you check if the proposed symbols are available in OmniData ... and please make sure you give me the specific characters that OData EOD uses for that symbol.

If it's not avail thru OData, sorry, I won't be able to support it for now.
Top of the page Bottom of the page
Jump to page : 1
Now viewing page 1 [50 msgs/pg]
( E-mail a Link | Printer Version | Search Room )

Owner of site: Jim Dean -- Forum content is confidential, and may not be distributed without written permission.