Pages: << 1 2 3 4 5 6 7 8 9 >>


Permalink 09:51:31 am, by Dana Comolli Email , 646 words   English (US)
Categories: Reconciliation, Accounting

Accounting from broker statements; not such a great idea

This sounds like a great idea: generate accounting and performance numbers for managed accounts and funds by reading broker statements.

On the surface, it is a great idea; unfortunately, it is not likely to succeed in the context of a CTA or hedge fund that trades futures and/or FX; not to mention over the counter instruments such as swaps.

There are several reasons why this can’t work reliably. The most notable are:

  • Traders of any size have out trades every day. Basing accounting on the contents of broker statements that have missed or incorrectly booked trades is guaranteed to be wrong.
  • Broker statements for futures are cash based, not accrual based. They do not include the accrued trading commissions for the exit side of the trades. If performance is based on the contents of the statement, that performance will be over-stated by the amount of the accrued exit commissions. The statements also do not contain accrued interest on cash held at the broker. Again, basing performance on the contents of the statements results in inaccurate performance calculations.
  • Statements do not include management or incentive fee accruals. These must be calculated outside the context of the data on a statement.

Another problem that believing what is on a statement causes is it defeats the whole concept of reconciliation. By accepting what is on a statement, the advisor is in effect, doing the same thing as never balancing a checkbook.

In the case of an investment manager, he is being paid to manage the money of his investors. It is the manager’s responsibility to ensure that trades are booked in the correct account at the correct price, that positions are accurate and balances are in line. Basing accounting on the contents of the statements bypasses these important checks.

A better approach
If you assume an investment manager must maintain the list of trades she has made as well as her open positions in order to effectively run her business, she is already most of the way to being able to produce the required financial performance records.

A system like TheBooks is designed to take advantage of this fact. It provides simplified recording and reporting of trades and automatically keeps track of positions (both net to the street as well as by system/strategy) and it generates the appropriate accounting entries resulting from those trades and open positions.

A clear advantage of this is that reconciliation can be performed at the trade, position, and cash balance levels (another capability of TheBooks). In addition, because commission, fee, and interest accruals are also performed by TheBooks, the resulting accounting and performance reporting numbers are accurate.

Another advantage to performing your own accounting rather than basing it on the contents of broker statements is time. If accounting/reporting is based on broker statements, there is no easy way to provide end of day performance reports. All reporting must wait until the statements are received; typically early the next day.

TheBooks reporting is based on the trades that have been posted and the market prices as they are now. This means that end of day reports detailing daily/monthly/annual performance (summarized or detailed by sector and/or market) can be sent automatically at the end of the trading day; significantly improving the level or reporting and transparency an advisor can provide to her clients.

An advisor must maintain a record of his trading activity. Using a system like TheBooks not only maintains a record of all trades, it maintains positions and generates all the performance and accounting records required in a way that delivers more accurate and timely reports than those that can be produced by using broker statements.

Beyond that, it provides the basis for robust trade, position, and cash balance reconciliation, helping the advisor manage the inevitable trade and position discrepancies that occur when trading futures and FX.


Permalink 11:17:32 am, by Dana Comolli Email , 926 words   English (US)
Categories: Data Management

Data Manager - The systematic trader's best friend

With version 2.2.1 of TheBooks, the Data Manager has come of age. This release has transformed the Data Manager from being simply an historic data repository and continuous contract builder into a research tool for futures traders which allows the combining and manipulation of data never seen before in commercial data management software.

TheBooks Data Manager Viewer

The 2.2.1 release of TheBooks adds the following features to the Data Manager:

  • The ability to define derived data sources. These are data streams which are based on other data streams (more about these later…)
  • The ability to overlay your actual trading history onto price charts. The running position and P&L can also be charted as part of this option. When trades are overlaid onto a continuous contract chart, the trade prices are adjusted to match how the continuous contract was adjusted.
  • Direct access to CSI data via their API. You no longer have to export data from Unfair Advantage to incorporate the data into TheBooks. You just tell TheBooks what computer in your network is running CSI and TheBooks is able to directly access the data.
  • The charting package has been completely re-written and is now a high-performance utility with the ability to quickly scroll through large data sets and zoom in and out allowing you see as much or a little detail as required. In addition, gaps in the data can be highlighted as can roll dates.
  • We have added the ability to easily switch between viewing any data set by simply clicking on the data set name within the viewer.
  • Export formats can be easily defined using a point and click interface. This includes the ability to translate market symbols from one symbol set to another.
  • For those traders that want tighter integration, TheBooks API (Application Programming Interface) has been extended to allow direct access to Data Manager data. This means your research platform and/or signal generator can read price series directly from TheBooks without the need to have the data exported to text files.

Even though DMAXX spent considerable effort on these enhancements, the features of the Data Manager that had set it apart from other data management applications have been retained and enhanced as well. These include:

  • The ability to use actual trading history as a roll rule when building continuous contracts.
  • The ability to define export sets which allows multiple data sources to be exported together to a single location in your network and to flag them to be automatically exported anytime any of the underlying data sources change.
  • Multiple price data vendor support. This includes CSI, CQG, and eSignal, in addition to user-defined text file imports. This means you can easily use data from the vendor(s) which make the most sense for your research and trading.
  • Flexible data download scheduling allows you to define as many scheduled downloads as you like. The data is downloaded in the background and is performed automatically. Continuous contracts and derived data sources are automatically updated and, as described above, any exports that are required are automatically performed as well.

Derived Data Sources
Derived Data Sources are an exciting new feature introduced in the 2.2.1 release of TheBooks. A Derived Data Source is a Data Source whose underlying data comes from one or more other Historic, Continuous Contract, or Derived Data sources combined using user-defined or pre-defined transformations. Once a Derived Data Source has been constructed, it can be used to like any historic data source or continuous contract.

There are four types of Derived Data Sources:

A Ratio data source is simply the ratio of two underlying data sources. The numerator bar is divided by the denominator bar. This can be used to provide a relative strength data stream between the two instruments or can be used to create FX cross rates from two data streams of FX rates.

An Index data source is the weighted sum of all the underlying data sources. For example, a user could define an Index-type derived data source as being a basket the major long-term interest rate futures contracts (US and foreign) and weight each as desired. The data stream would be the weighted sum of each of the underlying stream’s Open, High, Low, and Close. The trader could then track the index and Buy/Sell the basket of contracts as appropriate.

Term-Infused Spot
A Term-Infused Spot data source is a way to insert the term structure of continuous contract futures data into a data stream of daily spot prices. This provides a way to produce both outright and cross-rate historic price data that contains both a valid daily price range and term structure where currently the historic data does not exist. It uses an Historic Data Source which is an outright or cross-rate spot rate and a Continuous Contract.

A user-defined data source is a way for the user to apply his/her own transformations between two or more data sources. These transformations can include logical as well as numeric expressions, allowing to you perform operations such as splicing two data streams together at a specific date or comparing data streams from multiple sources and using the values from the stream that make the most sense on a given date within the data.

The Data Manager has evolved into a premier data management tool for the professional futures trader and researcher. These new features and those to come are designed to facilitate better research and better trading. It you are not using the data manager in your operation, now is the time to review its capabilities again.


Permalink 05:26:16 pm, by Dana Comolli Email , 564 words   English (US)
Categories: Disaster Recovery, Virtual Machines

Virtual Machines

A subject which we have getting a lot of questions about is the use of virtual machines. We at DMAXX have been using virtual machines in our operation for more than three years. In fact, we are so dependant on them that virtually everything in our network is a virtual machine. It is safe to say we could no longer operate effectively without them.

We use VMWare in our operation and have effectively virtualized all of our systems with the exception of our primary domain controllers and our SQL Servers.

Our use of virtual machines can be grouped into three major categories:

  • Development/Testing
    Because we usually have at least two and sometimes three versions of TheBooks in the field at once, having separate build and test systems for each version is a must. Virtual machines allow us to maintain these systems within the context of a single large server without the need to maintain separate systems for each version.

    In addition, installation testing is much easier as starting with fresh version of an operating system is simply a matter of copying some files.

  • Production
    The use of virtual machines allows us to segment functionality within our network to the point where each system performs a single function. This makes maintenance and upgrades much easier. It also allows the movement of servers from one physical computing platform to another by simply moving the files which represent the virtual machine.

    Hardware upgrades become much easier as upgrading a single platform results in the upgrade of all virtual machines resident on that system.

  • Experimentation
    Since we have to stay on top of the latest releases from Microsoft, we need to test our software prior to the release of their new operating systems and platforms. By using virtual machines, we are able to configure a standalone system with the new software and test it without going through the trouble of loading a physical system with the new operating components. Once we are finished, we can simply delete the virtual machine.

Disaster Recover
Another huge benefit of the virtual machine environment is in the disaster recovery area. Because the functionality within the network becomes portable, the ability to implement remote sites becomes greatly simplified.

Each of our VMWare host servers have backup scripts which iterate their inventory of virtual machines and one-by-one shuts down a VM backs up its configuration (virtual disks, hardware configuration, etc) to an NAS storage array and then re-starts the VM.

The NAS storage array is replicated to a remote location which also houses VMWare servers. If the primary site goes down, the replicated virtual machines can be started at the remote locating in a matter of a few minutes.

Virtual Machines and TheBooks
Needless it say, TheBooks will operate just fine within a virtual machine environment; all of our testing is done that way. We do, however, suggest the following guidelines:

  • SQL Server should be running “on the metal” and not within a virtual machine.
  • You can host your virtual machine on the same system that is hosting SQL Server. We would not recommend this for higher volume environments, but for smaller advisors or those with relatively low trading volume or small numbers of accounts, this will work fine.
  • For higher-volume environments, we would suggest hosting the virtual machine which is running TheBooks on a server which is separate from the SQL Server.

<< 1 2 3 4 5 6 7 8 9 >>

September 2020
Mon Tue Wed Thu Fri Sat Sun
 << <   > >>
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30        
The official blog for users and others interested in TheBooks®


XML Feeds

powered by b2evolution free blog software