prosper — Prosper user loans (2005-2011)

Description

A network of loans between the members of Prosper.com, an online peer-to-peer lending marketplace, from November 2005 to September 2011, compiled by U. Redmond and P. Cunningham. On Prosper, borrowers post listings requesting loans, and lenders bid on them; a listing becomes a loan if the requested amount is fully funded by the lenders. Each directed edge goes from a lender to a borrower, and represents the participation of the lender in a loan to the borrower; a single loan therefore appears as several edges, one per lender. Repeated edges between the same lender and borrower are kept. Users are identified only by anonymous numbers.

The network has the following properties:

(Edge) time: the date when the loan originated, as Unix time (at the resolution of days).
(Edge) amount: an amount in US dollars (91% of the values are between 25 and 100 dollars, which suggests that it is the amount lent by the lender rather than the amount requested by the borrower).
(Edge) status: the status of the loan when the data were collected (Paid, Current, Charge-off, Defaulted, Late, Payoff, Repurchased, Cancelled, and the undocumented values 1, 2, 3 and 4+).
(Edge) lender_rate and borrower_rate: the interest rates of the lender and of the borrower, as fractions.
(Edge) grade: the credit grade of the borrower (AA, A to E, HR for high risk, and NC for no credit score).

Tags

Social Online Economic Weighted Multigraph Timestamps Metadata

Added

2020-08-11

Citation

U. Redmond, P. Cunningham, "A temporal network analysis reveals the unprofitability of arbitrage in The Prosper Marketplace." Expert Systems with Applications 40(9), 3715-3721 (2013) 10.1016/j.eswa.2012.12.077 Sci-Hub

Upstream URL OK

http://mlg.ucd.ie/datasets/prosper.html

Networks

Tip: hover over a column header for a description.

Column legend

$\left<k\right>$: Mean degree
$\sigma_k$: Standard deviation of the degrees
$\lambda_h$: Largest eigenvalue of the Hashimoto (non-backtracking) matrix
$\tau$: Random walk mixing time
$r$: Degree assortativity
$c$: Global clustering coefficient
$\oslash$: (Pseudo-) diameter
$S$: Relative size of the largest component
NPs: Node properties
EPs: Edge properties

Name	Nodes	Edges	$\left<k\right>$	$\sigma_k$	$\lambda_h$	$\tau$	$r$	$c$	$\oslash$	$S$	Kind	Mode	NPs	EPs	gt	GraphML	GML	csv
prosper	89,269	3,394,979	38.03	145.96	327.23	5.28	-0.07	0.00	8	1.00	Directed	Unipartite		weight time	13.0 MiB	21.7 MiB	20.0 MiB	22.1 MiB

Ridiculograms^*

^* These are automatically generated force-directed visualizations, and can be quite meaningless for networks both big and small. They should not be taken seriously as sources of scientific insight. See here for a discussion.