| Revision as of 20:24, 13 October 2023 editRonnotel (talk | contribs)Extended confirmed users7,164 edits temp | Latest revision as of 20:29, 13 October 2023 edit undoRonnotel (talk | contribs)Extended confirmed users7,164 edits purgeTag: Blanking | ||
| Line 1: | Line 1: | ||
| The RTO in Chicago Wacker office has reached a new stage after the Oct 1st date. When we came back to the office 1.5 years ago, we were getting ~35 people on the Chicago risk floor on Tue-Thurs. That went up to the high 40s to low 50s this spring. The numbers have been steadily falling since then. Currently there are around 22 people on the mid-week days. We now have 7ish margin risk folks down from 11. | |||
| Most of the 2023 Margin Risk 1LOD Integration Issues that were bugs have been resolved or will be soon. There are a number of issues that aren't bugs with the first line tools. There are a few model questions (like how expiring options are handled or vol shock stuff) and a handful of user interface problems like responsiveness, filtering and settings stuff. We are in a better place system wise for NovTG, but the complexity of accounts is greatly increasing. | |||
| The one remaining bug is that we show incorrect data banners in SSC when PM customers break Portfolio Margin Alternate Calcs, but SSC is going to hide them. | |||
| We have a customer in Taiwan that is regularly short 100K options in NVDA, 22999999. She has assets > 600MM, and can take a -39% move in NVDA, but her position is too big. She has been set to closing only and 1LOD teams are discussing next steps with her. She's of particular interest because she's a driver for Option Clearing Corp CST scenarios. OCC is using Beta weighted tests at 25%, and so they could be stressing NVDA anywhere from 42% to 53%, depending on their beta assumptions. Hat tip to Jeff for looking up the monthly, dailly, and option implied betas. | |||
| Zilan led a meeting to describe her Oracle workflow vision with Paul, Stanley, and Quinn for AML. Ideally in the next two weeks, folks will get access to data, and be able to do analysis in early November. | |||
| John is working with a first line team to classify securities as Complex Products. The goal is to provide transparency to the customers that products like triple ETFs or crypto related investments have atypical risk profiles. He finished the prototype Shiny dashboard app (in R) that would manage a DB while creating an audit trail, doing data validation. There is a discussion about next steps and long-term support. | |||
| The October Prism release is on 10/17. Most of the changes have been documented here for MRO to understand but will also be in Definition of Done shortly. 2023 Margin Model Changes | |||
| Paul and STS got an AppDynamics running. This allows us to see detail in real time about transaction times, model runs, latency for the Prism systems. He has also built a healthcheck Splunk dashboard so that we can manage all of the new servers efficiently. He's been collaborating with STS for over 6 months on this effort. This was in Quinn's yearly goals to make the more complex MCM Cloud environment more maintainable. | |||
| JQL - we have confirmation from QP on moving forward with 2.0 for Blue | |||
| OVE | |||
| Most of the week was spent trying to setup OVE_DB (Dev) Access groups and testing - Dev DB looks good and we will get the prod db next week | |||
| Work Order for VM's was dormant since Sep 20, chased that down and they are going to try to get those setup next week as well. | |||
| Closed out conditional approval ticket for OVE Phase 1 approval was pending | |||
| Lost a day in an unseemly discussion with John K. about dashboarding tool. We would like to use Streamlit if possible but he wants to impose Shiny on everyone - unresolved | |||
| Katya is finishing up some QP performance documentation and Sean is still closing out the yield curve finding. JQL WP/TP updates were accepted and uploaded to Migraph | |||
| Sunil is working on parsing the big SVI params file published from MCM. Will need to help him out next week with that. | |||