How do you figure real time P/L?

[hlpsg]

If you trade options, what method do you use to figure your current P/L? Especially for deciding when to cut your losses and close the trade out?

For anyone trading spreads in the same month, any change in shape of the horizontal skew (between strikes) will affect your P/L, many times drastically. And this happens constantly throughout the day every few seconds.

Secondly, supply/demand at certain strikes which temporarily drive up/drive down IV of the options at those strikes will also affect your P/L drastically if you've bought/sold those strikes in your spread.

If you trade spreads which cross expiration months, it gets even more complicated with the intermonth skew.

So how do you solve this problem, and at any given time, how do you figure your P/L for the purpose of closing a trade because it has hit your max loss amount?

Thanks.

 

[ForexForex]

Setup a Yahoo Finance Portfolio and enter the trade including commissions and you can check the P/L without having to login to your trading account. The P/L will be within pennies of the real trade and then you can make changes to the real trade if necessary.

 

[spindr0]

The broker's portfolio page would be the simple answer as long as partial closing positions haven''t occurred (adjustments). That gets complicated because then you have to mentally track the P&L of the closed positions.

A more precise/accurate answer is that if your broker offers a DDE connection, you can customize an Excel spreadsheet to your needs and link it to real time.

 

[newwurldmn]

First of all changes in skew are legitimate changes in your pnl.

Secondly it will rarely change so much to have a substantial impact on your pnl unless you are running a very large portfolio of options.

If this is the case, then you will need to build your own paramterization model to flatten out the supply demand imbalances. (However, in your brokers' eyes you pnl will be whatever the change in skew is).

 

[hlpsg]

Quote from spindr0:

The broker's portfolio page would be the simple answer as long as partial closing positions haven''t occurred (adjustments). That gets complicated because then you have to mentally track the P&L of the closed positions.

Figuring past adjustments isn't really a problem for me.

However, my broker's portfolio page is dynamic and changes according to the vol skew so I still have the same problem.

Even if it were static I suspect it would just take a snapshot at any given time (which is unreliable) and doesn't really have a smoothing algorithm.

A more precise/accurate answer is that if your broker offers a DDE connection, you can customize an Excel spreadsheet to your needs and link it to real time.

Yes, I think this is the direction to go.

I'm wondering if there is an algo out there that helps to smooth out the curve of the horizontal vol curve.

What I want to do is to remove the "kinks" in the vol skew curve resulting from supply/demand imbalances at certain strikes, and secondly, hopefully find an average slope of the vol curve in the last X minutes, whatever makes sense.

Then I plug in these IV values from the hypothetical vol curve back into the options pricing model to derive a price, and use that instead of the real time P/L to figure out my hypothetical P/L if these imbalances didn't exist.

Need some help figuring out how to do these 2 things in math.

 

[hlpsg]

Quote from newwurldmn:

First of all changes in skew are legitimate changes in your pnl.

Yes I understand. However, I do not want to close out a position just because of temporary imbalances at certain strikes.

Secondly it will rarely change so much to have a substantial impact on your pnl unless you are running a very large portfolio of options.

I'm not trading very big, and yet these imbalances in the vol skew can typicallycan have an $800 - $1000 effect in my P/L.

If this is the case, then you will need to build your own paramterization model to flatten out the supply demand imbalances. (However, in your brokers' eyes you pnl will be whatever the change in skew is).

Great idea, yes I agree this is where I should go. I think using Excel and VBA might be a plausible solution for me.

Any books or resources I might look into to figure out how to do this?

Thanks very much everyone.

 

[spindr0]

Yes, figuring past adjustment totals isn't a problem but since the broker's portfolio page reflects only open positions, it's just something else to remember. To get around that, I list open positions as well as a line for booked profit/loss. Also figure total and individual gain per leg as well as some other calcs. None of this is mp't other than I see the info that I want.

I found that it's cleaner to use one SS for quotes and link another to it for my needs. That way, when you blow it up, you only have to rewrite 1/2 of it

As for your math problem, sorry, can't help you because it's above my pay grade.

 

[quatron]

The simplest thing to do is to fit a quadratic or cubic spline to the implied vols. You can use least squares optimizer to do this. All you need to do is to play with the spline parameters like the number of knots to get the desired effect. Then the resulting curve is your vol curve - back out theoretical prices from fitted volatilities and use them to calculate your PnL. Easy

 

[newwurldmn]

You will need to get live data into excel, and the back out the vol to create a raw vol surface. Then there are algorithms to smooth the surface. Largely this involves some kind of optimization problem fitting a cirv to the data points (like a non linear regression).

Livevol will sell you their parameterizations but when I looked at it last, it wasn't very good at the wing (many years ago).


I did it many years ago when I was a market maker to find those dislocations.

I think that the ilegitmate skew changes (on one strike) ar not as big of a deal as you think. Market makes move their whole curves as they largely see different strikes a fungible.

If your issue is that any skew change is illegitimate there are easier ways to solve for that.

 

[quatron]

Quote from newwurldmn:

I think that the ilegitmate skew changes (on one strike) ar not as big of a deal as you think. Market makes move their whole curves as they largely see different strikes a fungible.
[/B]

Depends on the market. In some markets MMs don't have any control because the market is so deep and the skew becomes very kinky. MMs smooth the kinks by trading spreads when there's true arb opportunity but the actual vol curve looks very weird sometimes. E.g. this happens with KOSPI 200 quite often.