US 12,288,230 B2
System and method for analysis and presentation of used vehicle pricing data
Michael D. Swinson, Los Angeles, CA (US); Isaac Lemon Laughlin, Los Angeles, CA (US); Meghashyam Grama Ramanuja, Santa Monica, CA (US); Mikhail Semeniuk, Golden Valley, MN (US); and Xingchu Liu, Austin, TX (US)
Assigned to TrueCar, Inc., Santa Monica, CA (US)
Filed by TrueCar, Inc., Santa Monica, CA (US)
Filed on Apr. 26, 2023, as Appl. No. 18/307,642.
Application 18/307,642 is a continuation of application No. 17/843,426, filed on Jun. 17, 2022, granted, now 11,669,874.
Application 17/843,426 is a continuation of application No. 16/888,383, filed on May 29, 2020, granted, now 11,392,999, issued on Jun. 29, 2020.
Application 16/888,383 is a continuation of application No. 16/148,695, filed on Oct. 1, 2018, granted, now 10,733,639, issued on Aug. 4, 2020.
Application 16/148,695 is a continuation of application No. 14/145,252, filed on Dec. 31, 2013, granted, now 10,108,989, issued on Oct. 23, 2018.
Application 14/145,252 is a continuation of application No. 13/554,743, filed on Jul. 20, 2012, granted, now 8,645,193, issued on Feb. 4, 2014.
Claims priority of provisional application 61/512,787, filed on Jul. 28, 2011.
Prior Publication US 2023/0259986 A1, Aug. 17, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 30/02 (2023.01); G06Q 10/06 (2023.01); G06Q 30/0201 (2023.01)
CPC G06Q 30/0278 (2013.01) [G06Q 10/06 (2013.01); G06Q 30/02 (2013.01); G06Q 30/0206 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for accurately estimating a price of a used vehicle over the Internet, the method comprising:
receiving, by a backend process executing on a processor of a vehicle data system, vehicle trim data, vehicle manufacturer pricing data, and vehicle transaction data from disparate data sources;
determining, by the backend process, vehicle residual values for corresponding vehicles across the vehicle trim data, the vehicle manufacturer pricing data, and the vehicle transaction data, wherein the vehicle residual values are determined by computing a function of an Exponential Decay estimate, a mileage, a condition, and a geographic region determined from the vehicle trim data, the vehicle manufacturer pricing data, and the vehicle transaction data for the corresponding vehicles, wherein the Exponential Decay estimate is determined by fitting an exponential decay curve to a depreciated valuation for each of the corresponding vehicles, the fitting resulting in a function, Y=βe−δt, wherein Y is a natural log of a vehicle's residual value as defined by sources that set residual values, β and δ are a first estimate and a second estimate from a non-linear fit regression model, respectively, e is Euler's number, t is time;
updating, by the backend process, a used vehicle transaction database with the vehicle residual values thus determined;
receiving, by a server computer from a user device via a user interface through the Internet, user input data about a used vehicle and geographic information including a zip code for the used vehicle;
responsive to the user input data about the used vehicle and the geographic information including the zip code for the used vehicle, constructing, by a frontend process executing on the processor of the vehicle data system, regression variables for a price ratio model, the regression variables representing features that impact an expected price ratio;
determining, by the frontend process, a subset of vehicle transactions stored in the used vehicle transaction database, the subset of the vehicle transactions corresponding to used vehicles having a year, make, model within a distance of the zip code, wherein the used vehicles are considered as being in same bin as the used vehicle, wherein each vehicle in the same bin has same or similar vehicle year, make, model, and trim as the used vehicle, and wherein each vehicle in the same bin is in a geographic region inclusive of the zip code of the used vehicle;
populating, by the frontend process, the regression variables for the price ratio model with corresponding values from the subset of the vehicle transactions;
plugging, by the frontend process, pre-calculated regression coefficients into the regression variables to obtain the expected price ratio; and
generating, by the frontend process utilizing the expected price ratio, an estimated price for the used vehicle in the zip code, given the depreciated valuation calculated by the backend process for each of the corresponding vehicles in the same bin as the used vehicle.