FundSvcs Community

Hey all! Cross posting this here as that I know some of you are not on the APRA listserv. Below is what I posted over there. We’ve all had issues with our respective wealth screening vendors when it comes to screening common names. Due to the subsequent false positives from those records, I wanted a way to identify how common a name is within my wealth screening batch. The applications of this are plentiful, whether it be extracting names from a batch to save on credits or using the score as a derived variable for modelling. Located here is a spreadsheet <https://drive.google.com/file/d/1NLNiuqnrP3NRUP3-7pVAB32J1cKA2ipL/view?usp=sharing> that I created towards something for those purposes. Using data from the US Social Security Database, I’ve given scores to reflect how common a given name is. Just a quick rundown of methods: - The list of names come from the US Social Security database - I then standardized the counts for each name and made the cutoff point when names were above average in the standardized score - The names that made the cutoff were ranked and the ranks come from the standardized score with 143.76 being the highest ranked score (thus most common) I’m posting it here to get thoughts and ideas on how to make the document better, so feel to leave a comment on the document or get in contact with me if you would like to chat more about it (fyi, Sheets didn’t follow all of my formatting from Excel so things may look a bit wonky) *Steve Grimes Director, Development Analytics and Strategy Jazz at Lincoln Center Venue / Frederick P. Rose Hall / Time Warner Center, 5th Floor Offices / 3 Columbus Circle, 12th Floor, New York, NY 10019 sgrimes@jazz.org <sgrimes@jazz.org> / jazz.org <http://jazz.org> P 212 258 9985 F 212 258 9900*

Steve, This is fantastic! Thanks for sharing it. I'll dig in. Cheers, Andy McMahon US Holocaust Memorial Museum On Mon, Jun 17, 2019 at 9:45 AM Steve Grimes <grimessjr@gmail.com> wrote: > Hey all! Cross posting this here as that I know some of you are not on the > APRA listserv. Below is what I posted over there. > > > > We’ve all had issues with our respective wealth screening vendors when it > comes to screening common names. Due to the subsequent false positives from > those records, I wanted a way to identify how common a name is within my > wealth screening batch. The applications of this are plentiful, whether it > be extracting names from a batch to save on credits or using the score as a > derived variable for modelling. > > > > Located here is a spreadsheet > <https://drive.google.com/file/d/1NLNiuqnrP3NRUP3-7pVAB32J1cKA2ipL/view?usp=sharing> that > I created towards something for those purposes. Using data from the US > Social Security Database, I’ve given scores to reflect how common a given > name is. > > > > Just a quick rundown of methods: > > - The list of names come from the US Social Security database > > - I then standardized the counts for each name and made the cutoff point > when names were above average in the standardized score > > - The names that made the cutoff were ranked and the ranks come from the > standardized score with 143.76 being the highest ranked score (thus most > common) > > > > I’m posting it here to get thoughts and ideas on how to make the document > better, so feel to leave a comment on the document or get in contact with > me if you would like to chat more about it (fyi, Sheets didn’t follow all > of my formatting from Excel so things may look a bit wonky) > > > > > > > > > *Steve Grimes Director, Development Analytics and Strategy Jazz at Lincoln > Center Venue / Frederick P. Rose Hall / Time Warner Center, 5th Floor > Offices / 3 Columbus Circle, 12th Floor, New York, NY 10019 > sgrimes@jazz.org <sgrimes@jazz.org> / jazz.org <http://jazz.org> P 212 258 > 9985 F 212 258 9900* > >