Forums  > Careers  > General advice  
     
Page 1 of 1
Display using:  

npcareeradvice


Total Posts: 4
Joined: Dec 2016
 
Posted: 2016-12-04 00:11
Hello

I am unsure where else I could turn with a question like the following:

I am approaching 7 years of career experience (5 years at a bank on a trading desk in a strat/quant role, 1.5 years at a vendor, and now at a large hedge fund/asset manager in a portfolio strategy/research role).

Since I came out of graduate school I have experimented with data and tested things I have found interesting and implemented those that have shown promise.

Since around 2013 I have been able to find things that work a bit better and have established a track record in a brokerage account on a systematic strategy annualizing around 33% per year (1 year trailing performance is 113%, yes better performance more recently. full period sharpe is around .9). The accounts are retail brokerage and use little to no leverage (2 of 3 accounts are retirement - full cash, one uses 20% margin)

Prior to joining the current job I shopped around what I am doing and was turned down at a number of places for various reasons (never managed money professionally, not long enough track record, maybe i am just poor at sales).

I am learning a great deal at current role and it is a dream place to be, yet I continue to want to run my own book and work on these ideas full time.

My expenses are low, I do not have children, and I love working to research investment ideas. The cost to run my strategy including retail t-cost is approximately 500$.

What should I do? Is there a route to go with this that I am not realizing? Should I sit tight at the good gig and have this PA trading as a nice bonus (I am within personal trading rules at job)?

I am sorry in advance if this comes off poorly. I trust this board and the wisdom of the people on it. I really have no idea what to do with these side projects and I feel the performance is pretty good.

tradeking


Total Posts: 7
Joined: May 2016
 
Posted: 2016-12-04 08:11
How scalable is your side strategy, and how correlated is it to common betas (stock market, momentum, etc)? It will be hard to find much institutional interest for a 0.9 sharpe strategy with a 3 year track record unless it can potentially scale into the hundreds of millions.

Is there any reason why you cannot extend this side research into your current role? It seems to be a good fit given that you are already in a portfolio research role. You would gain many more resources through the firm, and can potentially convince a PM to back some money on it (would give your strategy a lot more credibility for your next job search).

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2016-12-04 11:40
And as everybody will have understood by now, realized sharpe ratio is a poor predictor of expected future sharpe ratio.


If you are not living on the edge you are taking up too much space.

npcareeradvice


Total Posts: 4
Joined: Dec 2016
 
Posted: 2016-12-04 19:30
@tradeking:
If I stay within cash products 5% of 90 day trailing volume summed over assets I trade is 69M$ (Each product represents the same "view", so I believe sum makes more sense than mean here). I trade ~9x/year with hold time of approximately 30 calendar days.

I discussed this work while interviewing (I believe to my benefit). I will bring this back up at year end to see if it is worth bringing back up internally.

Average rolling 1 year correlation to ff factors:
Mkt-RF 0.46
SMB 0.23
HML 0.06
RMW -0.20
CMA -0.01
Momentum 0.00

@goldorak:
Sharpe is a commonly quoted statistic. If you want any other data points to offer input on I will happily provide.

tradeking


Total Posts: 7
Joined: May 2016
 
Posted: 2016-12-05 00:13
"I discussed this work while interviewing (I believe to my benefit). I will bring this back up at year end to see if it is worth bringing back up internally."

Bringing the strategy over to your work will probably be the best course of action. If you can use the resources of the firm to boost the sharpe up to 1.5-2, and scale up to a larger universe, you should have no problems getting interest from multi-manager platforms like Millennium, SAC, BAM, etc. in the future.

Until then, you cannot do much more than just trading this for yourself on the side (provided that the firm allows you to).

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2016-12-05 08:40
> Sharpe is a commonly quoted statistic.

To the benefit of a few.

If you are not living on the edge you are taking up too much space.

npcareeradvice


Total Posts: 4
Joined: Dec 2016
 
Posted: 2016-12-05 11:29
@tradeking thanks for input. Will keep at it. Yes, what I'm doing is fully allowed where I work.

@goldorak what would you want to see to evaluate things?

HitmanH


Total Posts: 413
Joined: Apr 2005
 
Posted: 2016-12-05 12:19
As a fan of Sharpe (trading delta one products only) - i'd agree - a Sharpe of sub 1 - no-one is going to be interested in (unless they know everything about it - and diversifies their other strategies)

I'd suggest trying to speed up the strategy (trading 9x a year is very low). Execution costs aren't that high - so see if you target say 26 turnovers (every 2 weeks) - you can get a better sharpe. Not always possible - but often a simple speeding up does help things...

That's a broad thing.

As for places to allocate capital - there are a few. Some places would want to see that you are running it FT; others not. Some will want to co-invest with you (maybe with you taking first / overweight losses). All depends...

EspressoLover


Total Posts: 206
Joined: Jan 2015
 
Posted: 2016-12-05 16:40
Re-calculate the historical performance if you had hedged out market beta. (Possibly sector and FF exposure as well.) S&P's been about 1.0 Sharpe since 2013, and you seem to be very long-beta biased. The first question any investor's going to ask is whether this is just glorified beta exposure.

I don't agree with the Sharpe hate. Sharpe's the worst strategy metric... except for all the rest. But the second question any investor's going to ask is what the drawdowns look like. Even if Sharpe's not great, if you can promise small drawdowns you'll get a lot more interest. That's because an investor can just set you up with a tight stop-loss, and mostly not worry too much about strategy internals.

Third question an investor's going to ask is how this would have performed in bear markets. 2013-2016 isn't exactly a diverse collection of market environments. Is it systematic? If so, can you backtest it going further back. If you can get 2008 in your performance, that'd be pretty informative. If you can get September 11, LTCM and the 90's tech bubble even more so.

+1 @Hitman. Speed it up. Your turnover's way too low. You're never going to get much better performance with your current holding periods. And that should be your primary goal. Having a 1 Sharpe strategy is like being the best player on your high school football team. Having a 2 Sharpe strategy is like getting signed to the NFL.

goldorak


Total Posts: 979
Joined: Nov 2004
 
Posted: 2016-12-05 18:48
Sharpe ratio, actually think of t-test, as any other metrics which can claim to be more efficient in the case of certain distribution, has two main issues.

First it is a single hypothesis test, not a multiple hypothesis test. Would you want to select an investment strategy based on this metric given the typical size of the opportunity set one is contemplating, you would need a realized (backtested is just way too easy to reach) sharpe well above the levels you would normally consider.

The second issue is its value as a forecasting factor for the future. Let's call it an R^2 of 0.0001 and close here the discussion?

It is not in the metrics, it is all in the right use of statistics and the right assesment of their forecasting power for t+1.

Unfortunately for the ugly world of finance, t+1 is everything.

ps: Not targeting EL and HitmanH here. I know that you know that I know that you know Evil Smile It is just that sometimes the new comers out there need to know the truth and confront reality. 15 years ago I would have easily posted the same. Just claiming a sharpe above 10...



If you are not living on the edge you are taking up too much space.

npcareeradvice


Total Posts: 4
Joined: Dec 2016
 
Posted: 2016-12-20 22:26
yes @goldorak, but unlike the younger you I have actually realized the sharpe.
Previous Thread :: Next Thread 
Page 1 of 1