Reputation
8,472
Next tag badge:
100/100 score
16/20 answers
Badges
23 62
Impact
~275k people reached

Aug
24
comment How do I find the most diversified portfolio, or least correlated subset, of stocks?
You could probably use DEOptim() R package to solve this complex objective function. Along the lines suggested, I would add a constraint for the max number of assets. Also, you can include in the objective function a vector corresponding to the expected returns of each stock. Since you are optimizing only over 10 stocks the algorithm would converge rapidly.
Aug
23
revised The T+H Problem in Factor model forecasts
deleted 2 characters in body
Aug
23
revised The T+H Problem in Factor model forecasts
deleted 138 characters in body
Aug
23
revised The T+H Problem in Factor model forecasts
deleted 138 characters in body
Aug
23
revised The T+H Problem in Factor model forecasts
added 52 characters in body
Aug
23
comment The T+H Problem in Factor model forecasts
Yes - I have a sliding window. The forecast for T+H is adjusted - let's say weekly. I don't see how that helps much - you still need to wait a full-year for the dependent variable to manifest itself in the training data.
Aug
23
answered The T+H Problem in Factor model forecasts
Aug
23
comment The T+H Problem in Factor model forecasts
a) In my case I am using daily data. b) weighting is a good idea to capture the changing correlation structure but it would have not worked in the above scenario -- a full one year after the bear market you would still be short. c) has the seeds of a good approach. There would be some method to calibrate returns to some factor that is measured each day. The devil is in the details though - how would you realize this specifically?
Aug
23
revised The T+H Problem in Factor model forecasts
edited title
Aug
23
asked The T+H Problem in Factor model forecasts
Aug
23
revised How to calculate optimal standard deviation bands for trading?
added 253 characters in body
Aug
23
answered How to calculate optimal standard deviation bands for trading?
Aug
23
answered How well does CAPM beta track the risk of a particular market relative to world markets?
Aug
23
revised How can I select the least correlated portfolio of assets?
added 10 characters in body
Aug
23
answered How can I select the least correlated portfolio of assets?
Aug
22
answered What is the best data structure/implementation for representing a time series?
Aug
18
revised Any known bugs with Yahoo Finance adjusted close data ?
deleted 1 characters in body
Aug
18
comment Why does this Co-integrated basket look too good to be true?
By marks I mean when the currency pairs are quoted. Since there is a USD on each contract, if these currency pairs trade co-terminously then my hypothesis is false. On the other hand if one currency pair's quote for the day closes ahead of another currency pair but they are both marked on the same date then you might have a spurious result that is simply based on calendar/time effects.
Aug
17
comment Why does this Co-integrated basket look too good to be true?
Since you are modeling currency pairs across the globe there might be time effects in the model. For example, one currency pair may contain information about another currency pair because of timings of when markets open and close but the marks all take place on the same date.
Aug
17
revised Which valuation measures are most useful for equity market timing?
added 1 characters in body