Reputation
3,591
Top tag
Next privilege 4,000 Rep.
Access 'trusted user' tools
Badges
9 16
Newest
 Yearling
Impact
~96k people reached

Feb
22
comment Monte Carlo based mean variance optimization
Wrt to your point about Monte Carlo extensions to MVO. Monte carlo simulations are also used for mean-CVar/ES optimization. However, that's not mean-variance. If someone gives you the mean and covariance, then they are clearly implying you don't need to deal with vine copulas (to pick on one thing in the Wikipedia section). Even if they did give you vine copulas, then you could still generate the distribution from Monte Carlo, calculate the mean and covariance of the simulations and pass that to the optimizer.
Feb
22
answered Why is a risk-free portfolio desirable?
Feb
22
comment Monte Carlo based mean variance optimization
The resampling procedure in that paper is the same that I was thinking of. I wouldn't stress about it too much: would you really want to work for someone who can't make their interview questions clear?
Feb
22
comment Monte Carlo based mean variance optimization
The way you've put it, I agree that it is poorly phrased. Even if you sampled the returns with Monte Carlo, you could still pass the expected returns and covariances of that result to any portfolio optimizer. They could have meant Michaud resampling, but just saying MC-based MVO is too vague.
Feb
20
comment Fama-Macbeth second step confusion
@ppidosaurus I think it should be standard error of lamda instead of standard deviation.
Feb
20
comment Fama-Macbeth second step confusion
@Freddorick You say "They are the same for each cross-sectional regression". I thought there are two ways to do it. You could either estimate the time series regression over the whole period once or you could do rolling regressions. I thought the rolling regression approach was more common, TBH.
Feb
1
comment Is it possible to deal with non-normal distribution in Black-Litterman model?
@vanguard2k Suppose MVN(u, sigma) represents the multivariate normal distribution and MVLN(u*, sigma*) is what you get if you transform it to a multivariate log normal. Both u* and sigma* depend on u. This makes writing the optimization problem more challenging.
Jan
29
comment Is it possible to deal with non-normal distribution in Black-Litterman model?
@vanguard2k I've tried to do the reverse-engineer assuming that prices were multivariate log normal instead of multivariate normal. I wasn't that happy with it. I think it can't hurt to think about what is an appropriate prior, but once you start mixing in all kinds of different assets (equities, fixed income, options), the reverse engineer method of constructing the prior becomes a bit unwieldy I think.
Jan
11
comment How to implement dummy variables into GARCH(1,1) model from structural breaks (ICSS)
He could write his own log-likelihood function for the Garch and then use Matlab to optimize.
Dec
16
comment Confusion on stationarity vs deterministic trend
I'm not sure your point about detrending a difference stationary process is that clear.
Dec
3
comment Estimate Beta of CAPM from Implied Volatility?
There's a number of papers on using option-implied betas to explain the stock returns.
Nov
11
comment Intuitive explanation of stochastic portfolio theory
@Richard I wasn't familiar with this work. Thanks for at least pointing me in the direction to it.
Nov
3
comment How to compute simple and log portfolio returns?
@slaw You're right about asset log returns not equaling the portfolio log returns. But it's a feature, not a bug. Log returns are easy to work with through time, not by cross-section. You have to convert asset log returns to arithmetic to calculate the arithmetic portfolio return. For rebalancing, I can't recall any reference or resource. Like I said, just think it through. It is too basic for this site to bother with. The investopedia article is here: investopedia.com/walkthrough/corporate-finance/4/return-risk/…
Nov
3
comment How to compute simple and log portfolio returns?
@slaw You haven't made clear what is wrong with the Zivot notes. If you want other sources, any college corporate finance textbook will have the calculation as well. I found it on investopedia also. For rebalancing, you really just need to spend a momentum thinking about it. Log returns are a little trickier. You need to calculate the portfolio arithmetic return first. If you need portfolio log returns you can convert after.
Oct
20
comment Variable becomes more significant when more variables are included
This is probably more appropriate for stats.stackexchange.com
Oct
16
comment How to estimate the beta of corporations?
Damodaran has a lot of stuff on his website too.
Oct
14
comment Annualized Sharpe Ratio calculation
I would consider @bushmanov's 64 ways to calculate SR to be the most useful point so far. Who really cares if you don't get his number exactly?
Oct
14
comment Annualized Sharpe Ratio calculation
Try: calculate the CAGR of the index, then subtract the average risk-free rate (not logged or de-annualized), and divide by the daily arithmetic (not log) standard deviation times the square root of 250 or 252.
Oct
13
revised French and Fama Three Factor Model - What is the correct formula?
added 18 characters in body
Oct
13
comment French and Fama Three Factor Model - What is the correct formula?
Because I forgot! Now fixed.