Timeline for Fast Algorithms for sum of independent random variables
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 18, 2018 at 12:41 | comment | added | Iosif Pinelis | When people try to answer your questions, it is usually impossible for them to read your mind. If you had a multi-peak condition in mind, but did not not state it in your question, it is better to assess the answer to your question as it was stated, and then possibly pose a new question with whatever additional conditions you may want to introduce. | |
| May 4, 2018 at 5:52 | comment | added | iceage3t | Also, the tricky part is that in my Quant finance model, those Xi represents cashflow for each term, varies from 5k to 200k. My numerical experiment shows the portfolio(sum of those independent r.v's ) distribution follows a multi-peak pattern.I not sure they can be approximated by this c*Z. Regards Bing | |
| May 4, 2018 at 5:47 | comment | added | iceage3t | I really appreciate your answer.Although I did not learn this Edgeworth expansion before, I guess the rough idea before your approximation is to replace Sn with the 'magnified compensated Poission process ' c*Z.My question is how do we know this single r.v is good approximation ,of course, we already know that the first 3 moments are matched, but we still have higher moments unmatched right? | |
| May 3, 2018 at 16:15 | history | edited | Iosif Pinelis | CC BY-SA 4.0 |
added 513 characters in body
|
| May 3, 2018 at 14:23 | history | answered | Iosif Pinelis | CC BY-SA 4.0 |