2

First, there are a few things I'm not clear about, like what the 'risk free' return is.. is there even such a thing in trading? or how to handle inactive days, etc.

Let's assume I have a period of 30 days. During these 30 days, I have 5 trades with their duration:

  • +5%, 2 days
  • -3%, 5 days
  • +1%, 3 days
  • +2%, 1 day
  • -2%, 3 days

I have 16 days where there is no activity.

How would I calculate the Sharpe ratio from these values? I see a lot of questions on the topic, but I'm looking for a down-to-earth "here is to calculate it based on this concrete data" rather than a formula that has elements that I don't know how to interpret in this specific context.

Thomas
  • 141
  • 2
  • No can do. You need daily returns for all days in the period for which Sharpe Ratio is wanted. – nbbo2 May 24 '22 at 00:03
  • does it make sense then to separate days with trading but no trades (profit 0) from days 'off', like holidays, etc? and just subtract the days off from the total days? – Thomas May 24 '22 at 00:10
  • If you have no position on some trading days, then for those days the return would be 0. Holidays typically are not included in the calculation. – nbbo2 May 24 '22 at 00:16
  • Ok got it. How would I do the calculations with the example data above then? I’m still not clear what is the risk free part of the equation for example – Thomas May 24 '22 at 00:21
  • You can minus the t-bill rate for the risk free part – Ralph Winters May 24 '22 at 18:20
  • What’s a little bit difficult for me here is that I’m dealing with crypto trading where quite a lot of things are different so much of the terminology and standards are lost on me – Thomas May 25 '22 at 16:00

1 Answers1

0

Firstly -- you need a value for each period during your trading window to calculate the Sharpe ratio. Looking at your data, you have three options IMO:

  1. Put 0.0 values for the days that you didn't trade
  2. Track your total portfolio performance, per day.
  3. Track the performance of each position, per day, and reconcile at the end

#1 would be a little odd if you're concern is individual position tracking.

#2 would be the easiest, but wouldn't really take into account individual trades which might run against what you're trying to achieve here.

#3 sounds great, but would be more complex to reconcile and report over a period of time. For example, if you calculate the SR for a position you held for 3 days, then again for a position you held for 5 days, the you effectively have 2 SR values over two differing periods (3 and 5 days, respectively.)

So how do you reconcile that? Straight mean? Weighted mean? What about the days you didn't trade? Are you negative on those days if the market moves up?

alphazwest
  • 101