<t>P&L analysis is only one part of this. Market-impact, transitory and permanent, order-book replacement dynamics, spreads, trading costs (slippage, etc.) all form part of this analysis. There may be other aspects to look at, but is also depends on what type of trading and asset-class. The more...

<t>DCFC: Maybe I should clarify my previous comment; my experience is that one a day-to-day basis you might be struggling with a particular aspect of Levy processes, or whatever. My boss doesn't really care; he cares about the end result and wants to see the simulations based on this or that or thos...

<r>You can draw your covarinace matrices from a Wishart distribution; there are some papers floating around where the authors have modelled covaraince matrices as a Wishart process. <URL url="http://www.wilmott.co.uk/messageview.cfm?catid=4&threadid=20836"><LINK_TEXT text="http://www.wilmott.co....

<t>Where I work it's everybody for him(her)self. I'm part of a team, well in principle, but in reality nobody really takes note of what I do on a day-to-day basis. Informally, there is a network of quants, but not officially. One thing I realised when I started working for a bank is that banks don't...

<t>Data cleaning/smoothing, call it what you will is an art, not science. Assuming you come up with a cleaning algorithm which is able to maintain microstructural information and gets rid of artefacts, you still face a problem in live-trading. Your trading algorithm will inevitably see the "noisy" d...

We're using Python for some of our analysis into highfrequency data; however all production is still in C++.

<t>C is not really used at all; C++ seems to be the preferred language. Whether that is justified or not is a different question. IT certainly have a tendency to over-engineer everything.As to whether you can choose which language you use; that depends on which desk you work. Our prop-desk has quant...

Robert Almgren hass published some papers.

<t>Very good summary, though focused on equities. FX and Fixed Income trading are copying some ideas from the equities side, and vice-versa. Some banks are now offering execution algorithms in fx. The algorthms aren't the same since the information you get varies from market to market, but conceptua...

<t>C# and C++ each have their advantages. We work with tick-data across multiple asset-classes. C++ for data-handling and the math part; C# for the GUI and developing ideas quickly, but it's slowly converging into C++.We don't use MatLab or Excel/VBA; even MatLab had problems with the data-volume we...

VaR analysis and Monte-Carlo in fx...

<t>This might a somewhat silly question, but how does one estimate the covariance matrix? I know the formulae; however, whatever one measures depends heavily on the data that is used (particularly in terms of the look-back period), and the actual data set (i.e. daily, weekly, monthly returns). What ...

checkhttp://www.wilmott.co.uk/messageview.cfm?catid ... 20836where it was previously discussed.

<r>The below thesis is a good starting point for refs.<URL url="http://www.hec.unil.ch/cms_mbf/master_thesis/0307.pdfThe"><LINK_TEXT text="http://www.hec.unil.ch/cms_mbf/master_t ... 307.pdfThe">http://www.hec.unil.ch/cms_mbf/master_thesis/0307.pdfThe</LINK_TEXT></URL> idea of this strategy is actua...