sharetrader
Page 256 of 261 FirstFirst ... 156206246252253254255256257258259260 ... LastLast
Results 3,826 to 3,840 of 3908

Thread: Harmoney

  1. #3826
    Member
    Join Date
    Aug 2017
    Posts
    134

    Default

    Quote Originally Posted by myles View Post
    Had wanted to work out how to create a 'Fence' chart (finicky to get right) so thought I'd give it a go for this. Result below:

    Attachment 10054

    Not sure if there is enough detail to be meaningful? Would running it for individual grades (i.e. A1, A2) be useful? Let me know what you think.

    Perhaps there is a better way to show what you want - or it might be best to use a Pivot Table once you have the data and slice and dice as you want?
    Despite not requesting this data , the result of grade vs enquiry is interesting.

    Shows that grade is a far more important then # of enquiry in predicting a default. Hard to gleen much (as there may be confounders such the cohort mix), but it seems to suggest a 6+ defaults at A-C grade spikes default; D to E grades it doesn't seem to matter, and at F there appears to be an inverse relationship ie more enquiries results in less defaults. Fascinating!

  2. #3827
    Member
    Join Date
    Aug 2017
    Posts
    134

    Default

    Quote Originally Posted by Cool Bear View Post
    Myles,

    No, I meant the date the loan originates. So the cohorts are loans originating in 1h2015, 2h2015, 1h2016, 2h2016 etc.

    So that, for example, we can see out of all the loans originating in 1h2015, how many A, B,...F have been sold/charged off etc..
    X axis 1h2015A, 1h2015B...etc
    Y axis percentage of loans that are charged off.
    No time lapse, just bar charts.

    This is so that we can see the percentage charge off rates for older cohorts compared to younger cohorts.
    Yep thats exactly what requested.

    I'm guessing with your IT skills you don't need me to tell you. However I run cohort off a vlookup on column B "Date". Run a column on the combined export, and you should have all the data you need. Like the Fence post graph btw, very sweet presentation!

    Capture.JPG

  3. #3828
    yeah, nah
    Join Date
    Mar 2017
    Posts
    373

    Default

    Early release of basic chart set for comment:

    summary.pdf

    I'll remove or break this link when the final data set and summary are done.

    Looking for comments on the graphs etc. Will probably add to this over time, but this is just to get it out there for some initial comments so I can fix most of what needs fixing before the final data set comes together. Decided on pdf to keep it all together and for ease of viewing.

    Notes:
    • The charts are all based on the full set of data (except where stated), which means the default values are watered down a little due to the approx. 9 month gap between when a loan starts and when a default is flagged as having occurred (possibly ~7 months in the past). Do we just want to live with this or make some arbitrary cut off? (or run another set)
    • Some of the earlier loans, I feel, don't reflect the more recent loans i.e. there was some 'dodgy', 'poorly' graded loans in the early days. These will likely be inflating the more 'current' default rates. Should we consider dropping some of the earlier loans out (some are still current)?
    • Perhaps the above two cancel each other out to some extent?
    • Any time based default details is broken due to 'incorrect' use of the 'Last Payment Date' field so some caution needs to be taken around this type of data.
    • Dollar values of loans i.e. 'Outstanding Principal' etc., are of no value, since they are only for the particular portion of the unique loan record in the data set - so it needs to be understood that using these values would result in meaningless detail. (A couple of values like the 'Total Loan Value' can be used - ratios might be okay.)
    • I've limited quite a few data sets to a population of at least 10, just to remove outliers so core detail in charts etc. aren't influenced.
    Last edited by myles; 12-10-2018 at 12:33 PM. Reason: Changed to 'raw' link to avoid dropbox interface.

  4. #3829
    Junior Member
    Join Date
    Jan 2018
    Posts
    26

    Default

    This is great Myles- awesome work putting it all together.

    Can I get you to double-check the Rewrite graph? (Am I reading it correctly that 0/6610 have defaulted?)

    Included in the dataset should be this loan, which although it had 0 successful payments before being re-written, still got classified by Harmoney as a Re-write. So there should be at least 1 on that graph? (It should have been included in the orders.csv uploaded by 'Wilma@Bedrock.Flinstones', yesterday afternoon)

    LAI-00091286.jpg



    **edit**

    I've gone and looked at the csv, and it looks as though Harmoney have adjusted the 'Previous Loan Pay-off (re-write) column to N/A, whereas in the Harmoney portal/dashboard under reports, it shows this loan was in fact a re-write. I get the feeling that the re-write info on charged off loans is unreliable, it looks to me as though the re-write info shown in the csv changes once the status goes from in-arrears -> charged off.

    LAI-00091286(csv).jpg
    Last edited by alundracloud; 12-10-2018 at 07:42 AM. Reason: **added new picture & text about re-written loans

  5. #3830
    Member
    Join Date
    May 2016
    Posts
    165

    Default

    Many thanks Myles. This is great. First thoughts after a brief squizz: 1. Some cohort based data would be helpful or maybe just exclude loans below a certain age eg 12 months. This could be at letter grade level if that helps. 2. Some measure of net return would be of interest though of course with changes in rates and algorithms there will be comparability issues and the time factor would need to be accommodated. Maybe apply to just completed loans? 3. When you say full data set I assume you are not including duplicates.

  6. #3831
    Member
    Join Date
    Sep 2012
    Location
    christchurch
    Posts
    337

    Default

    Quote Originally Posted by alundracloud View Post
    This is great Myles- awesome work putting it all together.

    Can I get you to double-check the Rewrite graph? (Am I reading it correctly that 0/6610 have defaulted?)

    Included in the dataset should be this loan, which although it had 0 successful payments before being re-written, still got classified by Harmoney as a Re-write. So there should be at least 1 on that graph? (It should have been included in the orders.csv uploaded by 'Wilma@Bedrock.Flinstones', yesterday afternoon)

    LAI-00091286.jpg



    **edit**

    I've gone and looked at the csv, and it looks as though Harmoney have adjusted the 'Previous Loan Pay-off (re-write) column to N/A, whereas in the Harmoney portal/dashboard under reports, it shows this loan was in fact a re-write. I get the feeling that the re-write info on charged off loans is unreliable, it looks to me as though the re-write info shown in the csv changes once the status goes from in-arrears -> charged off.

    LAI-00091286(csv).jpg
    I just had a quick look at my defaults in HM reports on the website. Of the last 10 charge-offs, 6 had indicated that they were re-writes but had 0 successful payments and 0 remaining payments. Maybe HM just "zeroize" the re-writes when loans are charge-off. So maybe the conclusion that re-writes are safer investments is not correct after all.

  7. #3832
    Member
    Join Date
    Sep 2012
    Location
    christchurch
    Posts
    337

    Default

    Quote Originally Posted by myles View Post
    Early release of basic chart set for comment:

    summary.pdf

    I'll remove or break this link when the final data set and summary are done.

    Looking for comments on the graphs etc. Will probably add to this over time, but this is just to get it out there for some initial comments so I can fix most of what needs fixing before the final data set comes together. Decided on pdf to keep it all together and for ease of viewing.

    Notes:
    • The charts are all based on the full set of data (except where stated), which means the default values are watered down a little due to the approx. 9 month gap between when a loan starts and when a default is flagged as having occurred (possibly ~7 months in the past). Do we just want to live with this or make some arbitrary cut off? (or run another set)
    • Some of the earlier loans, I feel, don't reflect the more recent loans i.e. there was some 'dodgy', 'poorly' graded loans in the early days. These will likely be inflating the more 'current' default rates. Should we consider dropping some of the earlier loans out (some are still current)?
    • Perhaps the above two cancel each other out to some extent?
    • Any time based default details is broken due to 'incorrect' use of the 'Last Payment Date' field so some caution needs to be taken around this type of data.
    • Dollar values of loans i.e. 'Outstanding Principal' etc., are of no value, since they are only for the particular portion of the unique loan record in the data set - so it needs to be understood that using these values would result in meaningless detail. (A couple of values like the 'Total Loan Value' can be used - ratios might be okay.)
    • I've limited quite a few data sets to a population of at least 10, just to remove outliers so core detail in charts etc. aren't influenced.
    Thanks Myles for taking so much time and effort in this. Fantastic work indeed!

  8. #3833
    yeah, nah
    Join Date
    Mar 2017
    Posts
    373

    Default

    Quote Originally Posted by Cool Bear View Post
    I just had a quick look at my defaults in HM reports on the website. Of the last 10 charge-offs, 6 had indicated that they were re-writes but had 0 successful payments and 0 remaining payments. Maybe HM just "zeroize" the re-writes when loans are charge-off. So maybe the conclusion that re-writes are safer investments is not correct after all.
    It was too good to be true. That's why I asked about it a little while back and when someone indicated that they had only invested in re-writes and had no defaults I wrongly took that as confirmation. Good to know, with confirmation now.

    Oh well, looks like this stat can go in the bin, as I can't see how a 'Charged Off' re-write can be differentiated from a 'Charged Off' non re-write...

    From what I've read, Harmoney have had a good few Software Developers/Coders influence their interface over time, I'm guessing they had a few hacks go through who don't know the value of data integrity

  9. #3834
    yeah, nah
    Join Date
    Mar 2017
    Posts
    373

    Default

    Quote Originally Posted by RMJH View Post
    1. Some cohort based data would be helpful or maybe just exclude loans below a certain age eg 12 months. This could be at letter grade level if that helps. 2. Some measure of net return would be of interest though of course with changes in rates and algorithms there will be comparability issues and the time factor would need to be accommodated. Maybe apply to just completed loans? 3. When you say full data set I assume you are not including duplicates.
    1. Next step I think - have to validate the basics first before getting deeper into it.
    2. Would have to think long and hard on this one - details like how much did 'you' invest in the loan, fee rate, tax rate etc. come into it, which this data set does not contain (individual data only - so it would be up to you to do that for your own data).
    3. Yep, I won't be reporting anything on the 'RAW' data (except total records), only the unique loan set is in the charts etc.

  10. #3835
    yeah, nah
    Join Date
    Mar 2017
    Posts
    373

    Default

    Ignore the 'Investment' chart. It is using one of those value columns that are at individual level which has no meaning in a combined data set (I use this on my loans, but no value here - I'll remove it).

  11. #3836
    Member
    Join Date
    Apr 2002
    Location
    , , New Zealand.
    Posts
    484

    Default

    Quote Originally Posted by myles View Post
    Looking for comments on the graphs etc.

    Great effort Myles. Don't understand your last graph though. Perhaps a short explanation would help here, something like HM estimates is roughly on the mark as actual deafults appear to average out at HM estimates etc.
    I think some more information on early repaid by loan grade and protect would be very valuable too. At the moment the aggregate paid off of this data set is 52.6%, but my own data shows Bs repaying much faster than average, and protect loans repaying much slower than average.


    Quote Originally Posted by myles View Post
    default values are watered down a little due to the approx. 9 month gap between when a loan starts and when a default is flagged as having occurred (possibly ~7 months in the past). Do we just want to live with this or make some arbitrary cut off? (or run another set)

    Perhaps cutoff at the date of most recent default, to include all chargeoffs regardless of how soon or late they were booked to individual accounts. This will exclude some early repayments but we can live with that, unless you want to produce a full set chart too to compare.


    Quote Originally Posted by myles View Post
    Some of the earlier loans, I feel, don't reflect the more recent loans i.e. there was some 'dodgy', 'poorly' graded loans in the early days. These will likely be inflating the more 'current' default rates. Should we consider dropping some of the earlier loans out (some are still current)?
    Perhaps the above two cancel each other out to some extent?

    If I were charting, I'd leave them in as grading evolution as well as application are subjective and will continue to change with time. Even if Harmoney has become smarter at vetting, the economy is still headed out of the Goldilocks zone, so leaving a buffer might not hurt. Charge-offs for this (lean in F, and maybe As) dataset stands at 4.8%, which is in the recent ballpark declared by HM


    Quote Originally Posted by myles View Post
    Dollar values of loans i.e. 'Outstanding Principal' etc., are of no value, since they are only for the particular portion of the unique loan record in the data set - so it needs to be understood that using these values would result in meaningless detail. (A couple of values like the 'Total Loan Value' can be used - ratios might be okay.)

    I think this is a progresssion of your earlier idea when you wanted to see how many had invested in a certain loan, to determine desirable loan characteristics in some sort of a covariant analysis. There is merit in aggregating the invested amounts per unique LAI, if this can be done easily. One can just follow the money then, to see what the contributors liked/invested in.


    Quote Originally Posted by myles View Post
    Oh well, looks like this stat can go in the bin, as I can't see how a 'Charged Off' re-write can be differentiated from a 'Charged Off' non re-write... From what I've read, Harmoney have had a good few Software Developers/Coders influence their interface over time, I'm guessing they had a few hacks go through who don't know the value of data integrity

    Amen. Well spotted alundracloud and Cool Bear. I also just had a quick look at my defaults in HM reports on the website. Of the last dozen charge-offs, half were re-writes with 0 successful payments and 0 remaining payments. So, HM is wiping off the re-writes in summaries, when loans are charged-off. This brings into question the sanctity of not just their initial listings data, but also the subsequent changes they make to it.
    Last edited by beacon; 12-10-2018 at 11:32 AM. Reason: rewrite info added

  12. #3837
    yeah, nah
    Join Date
    Mar 2017
    Posts
    373

    Default

    Just for info: Latest 'Charged Off'/'Debt Sold' loans in data set looks like:

    Month|Count
    -----------------

    2017-04|22
    2017-05|20
    2017-06|23
    2017-07|16
    2017-08|15
    2017-09|14
    2017-10|8
    2017-11|9
    2017-12|8
    2018-01|3
    2018-02|5
    2018-03|1
    2018-04|1

    So the latest one is 6 months old.

  13. #3838
    yeah, nah
    Join Date
    Mar 2017
    Posts
    373

    Default

    And to help complete the picture these are the 'Paid Off' numbers:

    Month|Count
    ---------------

    2017-04|364
    2017-05|420
    2017-06|325
    2017-07|310
    2017-08|345
    2017-09|303
    2017-10|295
    2017-11|282
    2017-12|243
    2018-01|263
    2018-02|203
    2018-03|171
    2018-04|73
    2018-05|49
    2018-06|10
    2018-07|9
    2018-08|3
    2018-09|7

    So a fair amount of early 'Paid Off' loans would be left out...hmm...

  14. #3839
    Member
    Join Date
    Apr 2002
    Location
    , , New Zealand.
    Posts
    484

    Default

    Quote Originally Posted by myles View Post
    And to help complete the picture these are the 'Paid Off' numbers: ...

    Month|Count
    ---------------

    2017-04|364
    2017-05|420
    2017-06|325
    2017-07|310
    2017-08|345
    2017-09|303
    2017-10|295
    2017-11|282
    2017-12|243
    2018-01|263
    2018-02|203
    2018-03|171
    2018-04|73
    2018-05|49
    2018-06|10
    2018-07|9
    2018-08|3
    2018-09|7

    So a fair amount of early 'Paid Off' loans would be left out...hmm...
    So, if I was worried about default rate being watered down, I'd include data upto April end/date of last default, because I'd want to include that last default for its effect on variate percentages. Cutoff at May end will get me the 50 odd early paid too. I'd definitely start data from the first loan on the books - given substantial numbers for both early repairs and defaults.

    And I'd do the same for the separate PP dataset beginning from the date of the first PP loan.

  15. #3840
    Member
    Join Date
    May 2016
    Posts
    165

    Default

    When you quote default rates, are these %'s by loan numbers or $ values ?

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •