an analysis of abc auction house demetria henderson regression final project department of...

Post on 01-Jan-2016

214 Views

Category:

Documents

1 Downloads

Preview:

Click to see full reader

TRANSCRIPT

An Analysis of ABC Auction House

Demetria Henderson

Regression Final Project

Department of Management

December 4, 2012

2

AgendaBackground

Purpose

Multiple Regression Variables

Statistical Analysis Methods

Results

Questions

3

BackgroundABC Auctions

• Wholesale vehicle auction in business for over 50 years.

• Over 20,000 employees in 100 locations worldwide.

• In North America, 75 auctions are in operation.• Handles approximately 500,000 vehicles.• Has approximately 11,000 full-time employees.• Does business with over 75,000 auto dealers.

• Generates annual revenues of a billion dollars.

4

Purpose

To determine which variables; performance, employee statistics, and safety variables, predict the profitability of ABC

Auctions by using multiple regression analysis.

5

VariablesObservations• N = 75 (record for each NA location)

Dependent Variable• Profitability (Earnings as percentage of revenue)

Independent Variables• Legacy• Full-time turnover rate• Part-time turnover rate• Number of cars sold• Workman’s comp claims/10,000 cars• Lot damage per consigned cars• Occupational and Safety Health Administration (OSHA)

accidents/10,000 cars

6

Statistical Analysis Methods

All Possible Regressions Report

Multiple Regression Analysis

7

All Possible Regressions4 0.573452 8.039024 3.393803 ACFG 4 0.534053 8.402098 9.711158 ACEG 4 0.524857 8.484607 11.185708 ACDG 4 0.521444 8.515019 11.732858 ABCG 4 0.503544 8.672813 14.603107 AEFG 4 0.500734 8.697324 15.053679 ADFG 4 0.489046 8.798533 16.927683 ABFG 4 0.466447 8.991002 20.551236 ABDG 4 0.466405 8.991357 20.557976 ADEG 4 0.456464 9.074724 22.151919 ABEG

8

Results of Initial Model4 Variables Selected• Sold, lotdampercar, ohsapercar,ftturn

Summary• R2 = .5735• Data was normal• No multicollinearity• No apparent issues with dependence or equal

variance• I.V. statisticially significant except for number of

cars sold• Transformation of number of cars sold may be

required

9

Results of Final Model

4 Variables Selected

• Inverse(Sold), lotdampercar, ohsapercar, ftturn

Summary

• R2 = .5968• Data was normal• No multicollinearity• All I.V. statisticially significant• No apparent issues with dependence or equal variance

10

Results: Final Model

Profitability

51.27

– .13*ftturn

– 82079.59*inverse_sold

– 11.47*lotdampercar

– 1.11*oshapercar

11

Summary of Results

The model indicates that performance, employee turnover, and safety are all important in determining the profits for the ABC Auction locations.

12

Questions

top related