1,233 reputation
719
bio website
location Paris
age
visits member for 3 years, 9 months
seen Sep 30 at 20:39

Oct
2
comment Running a simple alpha estimation test for statistical significance of a signal
They way you ask your question makes me think that you are looking for somekind of bootstrap test. A related paper on that would be : Pairs Trading: Performance of a Relative-Value Arbitrage Rule Evan Gatev. Of course this will not actually tell you if you have found alpha. For this I think you should be looking at parametrical test.
Jul
29
comment Is it more accurate to analyze returns on a calendar day basis than a trading day basis?
Agree with @chrisaycock and JL344. It is definetly not standard but assuming friday to monday returns come from the same distribution as thursday to friday is a bit insane. (most people might be)
Jul
17
comment Multilayer Perceptron (Neural Network) for Time Series Prediction
@lehalle excellent answser
Jul
5
comment How to better understand trading signals?
The problem with Principle component analysis is that you moght lose information about what really drives your returns. Also as a rule of thumbs, before assuming that you have a non linear relationship make sure that a linear methods realy results in underfitting your problem. If you want to reduce the number of variables you could use shrinkage regression for example.
Jun
21
comment How does Cornish-Fisher VaR (aka modified VaR) scale with time?
Have you tried to do the calculation yourself? In the case of IID returns it shouldn't be too complicated. Write log returns of aggredated returns as the sum of individual log returns. Then calculate the skewness and Kurtosis of the sum.
Jun
10
comment How should I compute the Sharpe Ratio for mid-frequency pair trading strategy?
It also has to be done that way because almost everyone out there is doing it on daily returns. This enables clearer comparisons between strategies.
May
29
comment What are the steps to perform properly a risk factor analysis on a portfolio?
Are you really looking for a generaly accepted method? I don't think this exists for anything related to quant finance. I ask this there are probably people that know reference for this but would not post it as a "generaly accepted method".
May
29
comment Optimal execution and reinforcement learning
@SRKX Sorry you are right. +1
May
25
comment How to interpret results of Johansen Test?
@techpaisa you have to successively test, the first one is null: r=0 against r>0. in your case you reject at 99% if your test (12.88) >24.60. If this was true you would then test r=1 against r>1 : 2.46>12.97. if you reject that then you can conclude that r=2. You do not need to know the degree of freedom as the critical values are tabulated
May
21
comment How to model time series of illiquid stocks - 400 observations (transactions) per 8 hours?
This is not specific of illiquid securities but you won't get any single model that answers your question. You will need to ask yourself what is the final aim of your model. This will highly depend on what you want to do with it. If it's algo trading the first most important characteristic will be out sample predictive performance. If you want to build a pricer you will probably be more interested in the way it estimates it's volatility.
May
19
comment Measuring co-movement at non-constant intervals
Could you be clearer? Are you talking about irregularly spaced time series?
May
16
comment Is there a closed-form solution for the partial autocorrelation function of a Markov regime-switching process?
@John Thanks for the reference
May
15
comment Is there a closed-form solution for the partial autocorrelation function of a Markov regime-switching process?
@John: Any out-sample predictive ability? I haven't found them very usefull in the past.
May
14
comment Is there a closed-form solution for the partial autocorrelation function of a Markov regime-switching process?
+1 for simplicity. Are you planning on applying this model to financial time series prediction?
May
8
comment How to build a mean reverting basket?
@VictorP: Your question is very similar to the one asked just yesterday by Quant Guy. The paper I cite in my answer is an attempt to build such portfolios. There is also a post on the tr8dr blog related to that matter. You can find it by searching "cointegration clusters Tr8dr" on google.
Apr
13
comment Two prices pass the cointegration test but there is a trend. How to check stationarity?
You can include a constant and a deterministic trend in your initial regression. You then test for the presence of the trend (ie the significance of the parameter before the trend term).
Jun
30
comment Variable Selection in factor models
You should be concerned with correlation among your inputs or at least used technics such as ridge regression when estimating your model.
Jun
28
comment How random are financial data series?
@asmaier: I just started reading about these tests but I think you may not have enough usable data finance.
May
31
comment Linear combination of gaussian random variables
You can look at the wikipedia page on multivariate normal distribution. There is a paragraph on multivariate normality test.
May
9
comment Minimizing Correlation
Can you be more specific about Fabozzi?