2
$\begingroup$

Using quantitative finances techniques on limited information, how might one go about finding similar(highly correlated) assets whose public information is available? The only data offered on a list of funds are yesterday's Daily Unit Values alongside up/down arrows, and last quarter's average returns %. No historical data is offered, and can't find similar those funds' ratings elsewhere.

I plan to implement a daily scraper that logs in, catches the DUV prices and slowly begins building time series tables for each fund. What would be the next step? Is it inevitable that I must also keep updated time series on all publicly available equities and funds data as well?

Improve this question
$\endgroup$
5
  • $\begingroup$ Better? Let me reduce. $\endgroup$ – Marcos Mar 13 '12 at 16:02
  • $\begingroup$ I've reopened this given your edits. Bear in mind that this site is intended for people who do this for a living. $\endgroup$ – chrisaycock Mar 13 '12 at 16:24
  • $\begingroup$ @chrisaycock okay thanks. Pls feel free to edit/discuss down to worthwhile content. $\endgroup$ – Marcos Mar 13 '12 at 16:32
  • $\begingroup$ Are you trying to do performance attribution on this fund? This may be problematic, as they are a TAA fund and the exposures can be expected to change unpredictably over time. Without in-depth conversation with the managers, I think it is impossible to know what they are really doing. If you are familiar with the industry, though, you can often read between the lines in their public marketing messages. $\endgroup$ – Tal Fishman Mar 13 '12 at 16:42
  • $\begingroup$ I wish I knew how to read between their lines, but can safely say it's a company with zero remaining credibility and a history of class-action lawsuits for misleading and entrapping,selling inappropriate products to elders, etc. Even being able to see a TAA's fund current track record would help me a lot in steering allocation decisions until I can get out, but they leave little to public(customers'!) scrutiny. Hence my motive to compute daily fresh correlations as next-best thing. But you make a good point that enlightened me after more research,thanks. And the quant problem interests me. $\endgroup$ – Marcos Mar 13 '12 at 19:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.