616 reputation
36
bio website
location
age
visits member for 2 years, 6 months
seen Aug 16 '12 at 1:55

Jun
16
awarded  Yearling
Jun
16
awarded  Yearling
Aug
5
comment Why use a column database for tick/bar data?
One file per table is not unlimited. It's taxing to the operating system to keep track of so many open files, so in effect, you've just offloaded the task of indexing all these independent tables from the database to the operating system's file descriptor table. The row-oriented concept is not flawed; you're just not taking full advantage of it when you make all these separate tables.
Aug
5
comment Why use a column database for tick/bar data?
The approach of one table per symbol in essence is the column-oriented approach. Row-oriented databases just aren't designed to handle a huge number of tables as effectively as a huge number of rows in one table. They are built on the assumption that different tables in a database hold fundamentally different types of data, so any relations among different tables are ad hoc, and have to be joined up at query time. (The primary key is the row number, by the way. Better to think about what fundamentally makes the row unique rather than just assign an arbitrary autoincremented integer.)
Aug
4
comment Why use a column database for tick/bar data?
As to Q2: For a row database, I don't see any other viable option than one huge table with Symbol, Date, O, H, L, C, V, as columns. You'll need two indices built into this table: say a primary key index ordered first by Symbol and then by Date, and another index ordered first by Date and then by Symbol. Essentially, at some level, a column-oriented database will have to do something like this internally to make its operations reasonably efficient. But if you're not willing to have (or your row database can't handle) one huge table with a Symbol column, then you do need a column database.
Jul
31
comment What are the pros and cons of applying for a patent on a financial model or trading system?
Patents really need to be restricted to ideas that are original, non-obvious, and capable of being incarnated as a device with some degree of tangibility. It is absurd that regardless of any standards whatsoever, anyone can patent broad claims on anything from natural human genes that were not even invented to general or abstract mathematical ideas.
Jul
29
comment Is it more accurate to analyze returns on a calendar day basis than a trading day basis?
Just a note: the time scaling I propose here is in accordance with what was discussed here quant.stackexchange.com/questions/3667/… and here quant.stackexchange.com/questions/3646/… -- according to bilkent.edu.tr/~berument/jef01.pdf Mondays' returns do have the highest variance, but it does not appear to be higher than that of the other days of the week by a factor of $\sqrt 3$ as one would expect from the natural scaling of i.i.d. returns by calendar time.
Jul
29
comment Is it more accurate to analyze returns on a calendar day basis than a trading day basis?
I should have asked more accurately: the real question is not "Is it more accurate?" but "Would it dominate in terms of being able to make inferences from the data?"
Jul
29
asked Is it more accurate to analyze returns on a calendar day basis than a trading day basis?
Jul
29
comment Proof for non-positive semi-definite covariance matrix estimator
I think you're starting to get at something: in financial terms, if the number of observations is less than the number of securities whose returns are observed each time, then it is always possible (in retrospect, of course) to construct a portfolio that would have remained constant in value during that time, but, as you say, it can never have a negative variance, unless you can trade imaginary shares on the stock market!
Jul
27
comment Proof for non-positive semi-definite covariance matrix estimator
Yes, it is the same as the calculation in your comment, just summed out entrywise for each observation, although I did arrive at the rank inequality slightly differently. So you are looking for a financial/applied sort of intuitive way to see this both rigorously and obviously....
Jul
27
revised Proof for non-positive semi-definite covariance matrix estimator
added 308 characters in body
Jul
27
answered Proof for non-positive semi-definite covariance matrix estimator
Jul
27
comment What is the use of the Euler equation in the Ramsey growth model?
There is a second-order equation in the wiki article en.wikipedia.org/wiki/… -- it's just a guess on my part, but your model may have already incorporated a solution to that equation by the time you're doing the optimization you mention.
Jul
27
comment How to properly interpret accrued interest of bonds
You have clean and dirty mixed up.
Jul
12
awarded  Commentator
Jul
12
comment With there being such a high demand for electronic trading or just trading in general why are market hours so limited?
Not sure about that, but for stock markets, companies like to wait until the market closes before they come out with bad news, and in general, some traders also like to wait until the market closes before making certain kinds of deals. It's really more about managing perceptions than anything else because the processes that drive the evolution of market prices always continue uninterrupted whether there is trading or not.
Jul
9
comment Why should there be an equity risk premium?
@TalFishman Of course you're right in that companies owned privately are no doubt more likely to borrow privately, too, but in any case at lower than normal interest rates. I didn't mean to refer to the spread per se, but to the ability to borrow at rates only highly rated companies can in normal times. And I still feel that a higher than usual equity premium is being demanded by investors today in public and private markets.
Jul
7
comment Why should there be an equity risk premium?
Another indication of an elevated equity risk premium is the dried-up IPO market and the booming junk bond market. Because of the premium demanded by investors, companies are finding it too expensive (in terms of dilution) to issue equity, especially when they can issue junk debt for investment-grade interest rates.
Jul
6
answered Why should there be an equity risk premium?