network profile
| bio | website | |
|---|---|---|
| location | ||
| age | ||
| visits | member for | 1 year |
| seen | May 8 at 12:05 | |
| stats | profile views | 4 |
|
May 8 |
awarded | Popular Question |
|
Aug 25 |
awarded | Editor |
|
Aug 25 |
revised |
How do you calculate “excess returns”? added 637 characters in body |
|
Aug 25 |
comment |
How do you calculate “excess returns”? If I wanted to take overnight returns into account, it is suggested that I take yesterdays closing price (virtually buying at the last final moment before market closes) instead of today's opening price. With made-up yesterday's closing prices, the following calculation would then apply (see edited OP). If the compounded return of the S&P100 index was -0.22%, then underperformance would be (-3.18) - (-0.22) = -2.96. Am I doing this correct? |
|
Aug 25 |
awarded | Scholar |
|
Aug 25 |
accepted | How do you calculate “excess returns”? |
|
Aug 25 |
comment |
How do you calculate “excess returns”? I've asked the same question on an investment forum on Reddit. Completely different solutions are suggested there. Although I'm quite sure that you are right, would you be so kind to review the suggestions and confirm that your suggestion is indeed the way to go? It would really mean a lot to me since I need to know for sure and start running the simulation (my thesis is due in 2 weeks). In the meantime, I'm upvoting your replies. Have some karma :-) |
|
Aug 25 |
comment |
How do you calculate “excess returns”? I'm performing a simulation for my masters thesis. I have gathered the number of positive and negative tweets about all S&P100 companies for a period of 102 trading days. Each day, I buy and sell the stock which had the highest positive-negative tweet ratio in the previous 3 days. I want to see if such a trading strategy would "beat the market". So I buy when the market opens and sell when it closes. I do not hold stocks overnight. Could you please explain the "1 + the returns"? The actual total return would then be 1 + (-1.64) + (-2.12) + (-0.39) = 3.15 instead of 4.15? |
|
Aug 24 |
comment |
How do you calculate “excess returns”? Thanks for your reply, it clarifies the difference in market returns for me. On my returns you say: "Similarly, you should calculate your return from purchase (presumably at market open day 1) to sale (market close day 3)". But the situation is slightly different: on day 1, I bought and sold HPQ for a return of -1.64%. Then on day 2 and 3, I respectively bought and sold FDX and PEP. So I only held each stock for 1 day. I believe this is a different situation from where I would have bought 1 stock and held it for the 3 days. Shouldn't return be the sum here instead of average? |
|
Aug 24 |
asked | How do you calculate “excess returns”? |
|
May 15 |
awarded | Student |
|
May 15 |
awarded | Supporter |
|
May 15 |
comment |
Given series $A$ and a correlation, is it possible to randomly calculate a fitting series $B$? Although I can see how you got that from the OP, I do not restrict to integer values. |
|
May 15 |
comment |
Given series $A$ and a correlation, is it possible to randomly calculate a fitting series $B$? But without bounds, the returned sequence matching [1,2,3] could be as large as [10000,20000,30000], for instance. I believe the educational purpose would be somewhat lost there. The boundaries would be there to return a sequence af approximately the same order of magnitude as the given sequence. Would it be feasible to, within bounderies, return an approximation of the correlation given? In other words, if the returned sequence would correlate not to 0.9344 but 0.899, then that would also be acceptable. Perhaps a plus/minus 10% boundary could be set? |
|
May 15 |
asked | Given series $A$ and a correlation, is it possible to randomly calculate a fitting series $B$? |
