Reputation
Next tag badge:
27/100 score
11/20 answers
Badges
11 36
Impact
~93k people reached

1h
comment Is there a countably infinite Sigma-Algebra? Why?
As not all of the quants here are experts in measure theory I would add from a blog: "there are no σ-algebras that are infinitely countable. This means that any σ-algebra S is either finite (and is therefore just an algebra) or very 'BIG' in cardinality, in the sense that it is uncountable." taken from here: yaronhadad.com/…
1h
comment Estimating correlation using EWMA
What's the use of calculation correlation using yearly data? Will the data from 2008, 2009 as yearly data points help you to understand the dynamics of 2016? 2015? The world is changing ... a yearly frequency is too low in my opinion.
2d
comment Closed form solution of PDE of Option Price
Everything fine .. I just thought that the correct expression is "closed form solution". So I edited the question. I hope this is fine with you.
Jul
3
comment What's the best proffesional forex market data feed out there?
Could you please use full sentences and less/no abbreviations? And a full question title?
Jul
3
comment What is the best data structure/implementation for representing a time series in C#?
doesn't this better fit to stackoverflow?
Jul
2
comment Fractals indicator (Bill Williams) R Quantstrat
Would be great if your question were a bit more self-contained with less links ... then we don't have to click on them and we don't have to fear that they die someday .... ;)
Jul
2
comment Conversion of SPY prices to ES prices
For clarity: eMini S&P 500 futures (ES), SPY ETF taken from here zerohedge.com/article/spy-vs-es-who%E2%80%99s-leading-sp-500
Jul
2
comment Conversion of SPY prices to ES prices
Would you mind defining SPY and ES? It might be pretty clear but should be included for clarity.
Jul
1
comment good R package for vectorized option pricing
I stumbled upon this post: r-bloggers.com/… there they propose to use the compiler package which can improve speed under some circumstances ...
Jul
1
comment good R package for vectorized option pricing
I see ... this is different.
Jul
1
comment good R package for vectorized option pricing
what do you mean by vectorized? Getting price and delta with one function call? Or getting the price for a bunch of options with one call?
Jul
1
comment Joint probability distribution only measures product sets?
Right! I think this is what the author wants to say ..
Jun
23
comment Covariance between two stocks in a two-factor model
You write "the factor model is" and then two numbers. This is not a model. Do the 2 stocks "have" 2 different models? This is very unclear. Please format the question and insert the formulas that describe the models.
Jun
23
comment Options on Volatility Control Index
+1 beause I did not know that "Volatility Control Indices" exist. Who publishes them? Where? Thanks!
Jun
19
comment The Distribution of Future Stock Price
It is difficult to say what is meant if they do not state it explicetly. But the word "instantaneous drift" could point to the GBM setting - thus lognormal prices ( i.e. log returns are normal).
Jun
18
comment Seasonal patterns in financial markets (weekday effects)
I usually like looking at rather recent data just because I think that the world changes and that e.g. 2008-2014 is more relevant for today than periods before that. Thus looking at the link there seems to be quite a strong pattern showing reversal in the period 2006-2010. I would love to see more recent data too. But for this period the effect looks strong to me - although in general the article summarizes that the effect is not reliable. Maybe not at all times - maybe not in old days but today !? ;)
Jun
18
comment Seasonal patterns in financial markets (weekday effects)
Hi, thanks for the references. Do they also cover patters during the week?
Jun
18
comment Extracting Signal from Noisy Data
Yes ... in modern markets you can not expect large sources of arbitrage (small $\beta$). But if $\beta$ is small then speaking the the terms of your model the error dominates and you are likely to lose money.
Jun
16
comment Regression model syntax
$y$ in this case is $r_{t+1}+ \cdots +r_{t+H}$. In training the model you have to form sets of $y$ and corresponding $x$ (which are terms involving $D$). These will overlap if you increase $t$ by $1$ only - which should be ok as this usually happens in time-series regression.
Jun
15
comment Regression model syntax
The term $D_{t_i,k}$ are based on past returns before $t$. So one could say that they are independent and that $r_{t+1} + \cdots$ is "dependent". I would rather say that these are explanatory for the latter. Do you want to find out how that $D$ terms are calculated precisely?