Search found 35 matches

by nickmar
Tue Jan 30, 2007 3:08 pm
Forum: Data Providers and other non testing software
Topic: CSI Stock Data: Reliability of Start Dates?
Replies: 20
Views: 20447

Darn - thanks for the info Tim. The lack of a unique security identifier is unfortunate as Ticker symbols are often re-used.
by nickmar
Tue Jan 30, 2007 2:43 pm
Forum: Data Providers and other non testing software
Topic: CSI Stock Data: Reliability of Start Dates?
Replies: 20
Views: 20447

AFJ, Sluggo: Do either of you know how to extract the CUSIP or PermNo identifiers for each security in the CSI database? I wanted to perform an issue date lookup comparison with the following excellent sources of data in order to determine the extent of the problem: Field-Ritter dataset of company f...
by nickmar
Tue Jan 30, 2007 11:05 am
Forum: Testing and Simulation
Topic: Kat, Goldman and copulas
Replies: 8
Views: 5911

Kat has a new website: http://www.fundcreator.com/
by nickmar
Tue Dec 05, 2006 10:16 am
Forum: Testing and Simulation
Topic: Kat, Goldman and copulas
Replies: 8
Views: 5911

Excerpt from the Kat and Palaro paper: "We also see that the replication strategy is unable to replicate the three large losses that LCH reported in October 1987 (-22.52%), August 1998 (-11.45%) and April 2000 (-10.83%). Since these are clearly outliers, it is not surprising that the replication pro...
by nickmar
Wed Nov 15, 2006 9:18 am
Forum: Money Management
Topic: Dynamic Portfolio Selection
Replies: 28
Views: 34549

Our shop employs a systematic approach to market selection. The only real constraint is market liquidity. Each market is categorized by sub-sector (e.g. short-term US interest rates) and exposure to each sub-sector is capped. Back-testing such an approach was not trivial since ALL markets (both acti...
by nickmar
Wed Nov 01, 2006 9:58 am
Forum: Stocks
Topic: Systematic equity funds? CTA's that trade stocks?
Replies: 18
Views: 21873

Eric: you certainly bring up a good point regarding Rentec. They have already surpassed the $10B mark in assets under management (in less than a year) and they were up over 12% YTD through August. Rentec does employ a long-term long-only approach but I suspect that their goal is return enhancement r...
by nickmar
Mon Oct 30, 2006 8:42 pm
Forum: Stocks
Topic: Systematic equity funds? CTA's that trade stocks?
Replies: 18
Views: 21873

You would be surprised how important the marketability aspect is in the equity long/short space. I suspect that the due diligence teams at fund-of-funds feel more comfortable allocating to a Long/Short Manager who utililizes a bottom-up approach and is able to "tell a story" about every name in thei...
by nickmar
Thu Oct 26, 2006 1:26 pm
Forum: Testing and Simulation
Topic: What is an "Edge"?
Replies: 32
Views: 33857

It turns out I was completely wrong here . I ran a short 10-test simulation where this appeared to be true but this just goes to illustrate that small sample sizes can be dangerous. Random Entries do not have an E70-Ratio (short for 70 day Excursion Ratio) of 1.0, it is more like 0.88. The values f...
by nickmar
Fri Oct 13, 2006 2:14 pm
Forum: Data Providers and other non testing software
Topic: CSI Price Format CSI#529 DJ STOXX 50 Eurex
Replies: 15
Views: 16362

I couldn't agree more with Sluggo - it's a small price to pay for 100% accuracy. We have even encountered contract specification errors on the websites of actual Futures Exchanges (usually one of the Far East Exchanges).
by nickmar
Fri Oct 13, 2006 12:17 pm
Forum: Money Management
Topic: Length of Time to Reach 100% Equity Risk
Replies: 17
Views: 19114

In an attempt to keep risk relatively constant over time (due to the disparity between the number of available markets in the 70s and 80s vs. the number of available markets today), you may want to consider adjusting your position-sizing algorithm to incorporate the number of markets available for t...
by nickmar
Tue Oct 10, 2006 11:11 am
Forum: Stocks
Topic: Distributions of stocks
Replies: 6
Views: 12088

Assuming that a trend-follower has the ability to short stocks as well, survivorship bias would imply higher returns in live trading compared to back-test (probably higher risk-adjusted returns as well).
by nickmar
Thu Aug 17, 2006 4:55 pm
Forum: Testing and Simulation
Topic: Portfolio Optimization
Replies: 50
Views: 38789

Market selection is a very big piece of the puzzle.
by nickmar
Thu Oct 06, 2005 11:06 am
Forum: Money Management
Topic: Dynamic Portfolio Selection
Replies: 28
Views: 34549

Tim - in order to implement the type of system that Dean described without using distinct sector portfolios, I suspect that additional functionality will be needed in the area of group management. For instance, there is currently no way to use the Ranking function to rank the instruments of a specif...
by nickmar
Thu Oct 06, 2005 7:27 am
Forum: Money Management
Topic: Dynamic Portfolio Selection
Replies: 28
Views: 34549

I suspect that such a system can be implemented in the current iteration of TBB. Each sector would probably need to be represented as a seperate portfolio. The same system would then need to be applied to each sector portfolio and then combined into one meta system. The instrument ranking functions ...
by nickmar
Fri Jul 08, 2005 12:03 pm
Forum: Testing and Simulation
Topic: Exit Strategy and Profitability
Replies: 13
Views: 12006

Sluggo - I think that the M/E Ratio can be calculated using TBB on a daily basis from the "Daily Equity Log" log file. Simply import the file into MSExcel and divide the "Margin Equity" column by the "Total Equity" column. The only shortcoming here is that only the current margin (i.e. not historica...