Project Paper in Word 97-2003

Embed Size (px)

Citation preview

  • 8/2/2019 Project Paper in Word 97-2003

    1/45

    1 Error! No text of specified style in document. | ACT 412

    ACT 412

    FALL 2011

    ABC IN AUTO

    INDUSTRY FROM

    CUSTOMERPERSPECTIVE

    SYED BOKHARI

    N O R T H E A S T E R N I L L I N O I S U N I V E R S I T Y

  • 8/2/2019 Project Paper in Word 97-2003

    2/45

    Overview

    In this paper, I will develop hypothesis regarding the relationship of

    the market price of the car and the features presented (PRODUCTPARAMETERS) in it. For example the price of a car is directly related

    to the year it has been manufactured, the mileage it has on the

    odometer, the styling (body type, xle, le etc), number of doors, the

    exterior and interior color, seats (leather or fabric), transmission

    type, engine size, sunroof, certified & owner type.

    I will collect around 100 samples and will use REGRESSION

    ANALYSIS to develop an equation which will be able to predict theprice by almost 90% accuracy and its key important variables which

    customer should focus on before buying a car. Also the

    manufacturers can review these parameters in order to improve

    their product costing in order to become more competitive.

    2 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    3/45

    TABLE OF CONTENT

    N

    O

    TOPIC PAGE #

    1. PRODUCT COSTING AND ITS DIFFERENT MEANING 4(a

    )

    Framework for cost accounting and cost management 4

    (b

    )

    Activity-Based Costing Systems 5

    (c

    )

    Using ABC for Improving Cost Management &

    Profitability

    6

    2. ABC IN AUTOMOBILES 73. DATA COLLECTION & VARIABLES 84. HYPOTHESIS 135. REGRESSION ANALYSIS AND ITS RESULTS 156. CONCLUSION 167. REFERENCES 188. APPENDIX-complete regression analysis 19

    3 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    4/45

    1) PRODUCT COSTING & ITS DIFFERENT MEANINGS

    Product cost is the sum of the costs assigned to a product for a specific purpose. Different

    purposes can result in different measures of product cost.

    Pricing & and Product-mix decisions: For making decision about pricing & which

    product provides more profit or contribution margin, managers are interested in the

    total profitability of different products & therefore assigns costs incurred in business

    functions of the value chain to the different products.

    Contracting with government agencies: Government reimburses contractors on the

    basis of the cost of a product plus a pre specified margin of profit. Due to cost plus

    profit margin nature of the contract, government agencies provide detailed guidelines

    on the cost items they will allow and disallow. For example some agencies dont allow

    marketing, distribution, and customer service costs from the product cost and may also

    partially reimburse the R & D cost.

    Preparing financial statements for external reporting under GAAP: Under GAAP, only

    manufacturing costs can be assigned to inventories in the financial statements. For this

    product cost includes inventoriable cost (DM, DL & MOH).

    The above paragraph indicates that product cost measures ranges from a narrow set o

    costs for financial statements to a broader set of cost for reimbursement under a

    government contracts to a still broader set of costs for pricing and product mix decisions

    Hence different purposes results in the inclusion of different cost items of the value chain

    of business functions when product costs are calculated.

    (a) Framework for cost accounting and cost management:

    4 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    5/45

    Three features of cost accounting and cost management across a wide range o

    applications are:

    Calculating the cost of products, services and other cost objects: Costing system traces

    direct costs and allocates indirect costs to products. Activity based costing system is used

    to calculate the total costs and unit cost of product and services.

    Obtaining information for planning, control, performance evaluation: Budgeting is the most

    commonly used tool for planning and control. It forces managers to look ahead, translate

    strategy into plans, co ordinate and communicate, and to provide a benchmark fo

    evaluation. Managers compare actual results to plan performance and to understand why

    difference (variances) arises and to use this information as a feedback to promote learning

    and future improvements.

    Analyzing relevant information for making decision: For this managers must understand

    which revenues and costs to consider and which one to ignore. Accountants help

    managers identify what information is relevant and what is not. When making strategic

    decisions about which products to produce, managers must know how revenues and cost

    vary with changes in output levels. For this they need to distinguish between variable and

    fixed cost.

    (b) Activity-Based Costing Systems:

    ABC refines a costing system by identifying individual activities as the fundamental cost

    objects. An activity is an event, task, or unit of work with a specified purpose-for example

    designing products, setting up machines, operating machines, and distributing products

    ABC system first calculate the costs of individual activities and then assign costs to cost

    5 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    6/45

    objects such as products and services on the basis of the mix of activities needed to

    produce each product or service.

    Three guidelines for refining a costing system are:

    Direct cost tracing: Identify as many direct costs as is economically feasible, thereby

    minimizing the extent to which costs have to be allocated, rather than traced.

    Indirect-cost pools: Expand the number of indirect cost pools until each of these pools

    are more homogeneous. In a homogeneous cost pool, all of the costs have the same o

    a similar cause and effect relationship with a single cost driver that is used as a cos

    allocation base.

    Cost allocation bases: Use the cost driver as the cost allocation base for each

    homogeneous indirect cost pool.

    The logic of ABC system is twofold. First, structuring activity cost pools more finely with

    cost drivers for each activity cost pool as the cost allocation base leads to more accurate

    costing of activities. Second, allocating these costs to products by measuring the cost

    allocation bases of different activities used by different products leads to more accurate

    product costs. ABC systems commonly uses a cost hierarchy having four levels-output unit

    level costs, batch level costs, product sustaining costs, and facilities sustaining costs.

    (c) Using ABC for Improving Cost Management & Profitability:

    Activity-based management (ABM) is a method of management decision making that uses

    ABC information to improve customer satisfaction and profitability. It broadly includes

    decisions about pricing and product mix, how to reduce costs, how to improve processes

    and decisions relating to product design.

    6 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    7/45

    Pricing and Product Mix Decisions: ABC gives managers information about the costs of

    making and selling diverse products, which helps them making pricing and product mix

    decisions accurately.

    Cost Reduction and Process Improvement Decisions: Manufacturing and distribution

    personnel use ABC systems to focus on how and where to reduce costs. Managers set

    cost reduction targets in term of reducing the cost per unit of the cost allocation base in

    different activity areas. Management can evaluate whether particular non value added

    activities can be reduced or eliminated.

    Design Decisions: Management can evaluate how its current product and process

    designs affect activities and costs as a way of identifying new designs to reduce costs

    For example, design decisions that decrease complexity of the mold reduces cost of

    design, materials, labor, machine setups, machine operations, and mold cleaning and

    maintenance.

    Planning and Managing Activities: To be useful for planning, making decisions, and

    managing activities, company specify budgeted costs for activities and use budgeted

    costs rates to cost products. At the year end, budgeted cost and actual costs are

    compared to provide feedback n how activities were managed. As activities and

    processes are changed, new activity cost rates are calculated. At the end of the year

    adjustments must also be made for under-allocated or over-allocated indirect costs fo

    each activity area using described method.

    2. ABC IN AUTOMOBILES

    Auto industry is one of the world biggest industries and many companies are incorporating

    different theories to reduce their cost, increase efficiency, and improve brand name

    7 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    8/45

    recognition and customer value. Lean manufacturing, just in time (JIT), and total quality

    management (TQM) are living examples of achieving higher excellence in these fields

    Toyota is one of best companies when we talk about all these theories and practica

    implementations. By practically implementing all these theories successfully, their ca

    sales growth was doubled in USA in just a small time of 7 years. Today they are world

    largest producers of cars followed by GM and WV.

    Toyota Camry is one of the most popular family sedan car in USA, therefore I have

    selected this car to study what important parameters customer prefer when buying it, so

    that the company can look into their costing as I will be identifying key components from

    customer point of view (with the help of regression analysis). They can compare their

    current ABC costing techniques by reviewing the results of this paper and then evaluate

    how can they further improve the quality of the product, production, reliability, and its cost

    and what changes they can make in order to give a better product to their customers.

    3) DATA COLLECTION & VARIABLES

    Initially I collected 106 observations from www.cars.com. All these observations are for the

    used cars as I wanted to know how much the price varies when the car gets older so that

    Toyota can review these parameters and can do the necessarily changes in order to

    improve the price of a used car which can further increase their image, sales, revenues

    profit and customer perceived value. Following are the key variables.

    VARIABLES TYPEList price Dependant variable

    1 Year Produced Quantitative2 Distance from 60625 (avg & approx) Quantitative3 Mileage Quantitative

    8 TERMPAPER | ACT 412

    http://www.cars.com/http://www.cars.com/
  • 8/2/2019 Project Paper in Word 97-2003

    9/45

    4 BODY TYPEType LE Indicator variableType SE Indicator variableType XLE Indicator variableType others Indicator variable

    5 EXTERIOR COLORSExterior color dark Indicator variableExterior color Medium Indicator variableExterior color Light Indicator variable

    6 INTERIOR COLORSInterior color Dark Indicator variableInterior color Light Indicator variable

    7 SEATSLeather seats Indicator variable

    8 ENGINE TYPEEngine 2.4L I4 Indicator variableEngine 2.5L I4 Indicator variableEngine 3.5L V6 Indicator variable

    Engine HP 4 Cylinder Indicator variableEngine HP V6 Indicator variable

    9 TRANSMISSION TYPEAutomatic Indicator variableAutomatic 5 speed Indicator variableAutomatic 6 speed Indicator variable

    10 TOYOTA CERTIFIEDCertified yes Indicator variable

    11 SUNROOFSunroof yes Indicator variable

    12 OWNERSHIPDealer or Individual seller Indicator variable

    Following is the raw data collected from the website in an organized excel format, which

    will be used for regression analysis.

    S:NO

    PRICE

    YEARS

    USED

    MNTHS

    USED

    DIST

    ANCE

    STYLE(CE,

    DX,

    MIL

    EAGE

    BODY

    EXTERIORC

    OLOR

    INTERIORC

    OLOR

    LEATHERS

    EATS

    FUEL

    EN

    GINE

    TRANSMISSION

    DRIVETRAIN

    DOORC

    OUNT

    WHEEL

    BASE

    SELLER

    TYPE

    TO

    YOTA

    CERTIFIED S

    UN

    ROOF

    9 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    10/45

    LE

    ,LEV6,

    SE)

    (dealer,private

    ow

    ner)

    1$29,3

    15 0 368 SE 9450 S D D Y G

    2.5LI4

    A 6SP F 4

    109 D N Y

    2$28,9

    95 2 2730 SE 5000 S L D N G

    3.5LV6

    A 6SP F 4

    109 D N N

    3$28,9

    91 0 328 SE 8160 S D L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    4$27,9

    91 0 311 XLE 4431 S L L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    5$27,9

    91 0 311 XLE 6028 S L L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    6$26,9

    91 0 328 XLE 9,585 S L L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    7$26,9

    00 1 1525 LE 13,584 S L D Y G

    3.5LV6

    A 6SP F 4

    109 D N Y

    8$26,2

    65 1 1532 NA 28,616 S D L N G

    2.5LI4

    A 6SP F 4

    109 D N Y

    9$25,9

    91 0 328 XLE 7,787 S M L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    10$25,0

    90 0 317 LE 25,090 S D D N G

    2.5LI4

    A 6SP F 4

    109 D N N

    11$24,9

    91 1 1528 XLE 9,874 S D L Y G

    2.5LI4

    A 6SP F 4

    109 D Y Y

    12$24,9

    80 0 316 NA 11429 S D L N G

    3.5LV6

    A 6SP F 4

    109 D Y N

    13 $23,999 0 3 17 XLE 8135 S L L N G 2.5LI4 A 6SP F 4 109 D Y N

    1423995

    2 2726 XLE 22917 S D L Y G

    3.5LV6

    A 6SP F 4

    109 D N N

    1523891

    1 1524 LE 33180 S D D Y G

    3.5LV6

    A 6SP F 4

    109 D Y N

    1623877

    2 2768 NA S D L N G

    3.5LV6

    A 6SP F 4

    109 D Y N

    1723786

    2 2717 XLE 15229 S D L Y G

    3.5LV6

    A 6SP F 4

    109 D Y N

    1823599

    1 1512

    SE13073 S D D N G

    3.5LV6

    A 6SP F 4

    109 D N N

    1923588

    1 15

    2

    6 XLE 11935 S D D N G

    2.5L

    I4 AUTO F 4

    10

    9 D Y N

    20$23,4

    90 2 2713 XLE 35405 S L L N G

    3.5LV6

    A 6SP F 4

    109 D N N

    21$23,2

    95 3 3932 NA 24878 S D D N G

    3.5LV6

    A 6SP F 4

    109 D N N

    22$22,9

    95 2 2726 XLE 33404 S D L Y G

    3.5LV6

    A 6SP F 4

    109 D N N

    23 $22,9 2 27 2 XLE 48838 S D L Y G 3.5L A 6 F 4 10 D N N

    10 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    11/45

    95 6 V6 SP 9

    24$22,9

    91 4 5131 NA 8465 S D D N G

    3.5LV6

    A 6SP F 4

    109 D N N

    25$22,9

    91 2 2711 XLE 13660 S M L Y G

    3.5LV6

    A 6SP F 4

    109 D Y N

    26$22,9

    91 2 2711 XLE 27221 S D L Y G

    3.5LV6

    A 6SP F 4

    109 D Y N

    27

    $22,8

    88 2 27

    2

    9 XLE 31273 S L D Y G

    3.5L

    V6

    A 6

    SP F 4

    10

    9 D N Y

    28$22,7

    80 3 3932 NA 19939 S L L N G

    2.4LI4 AUTO F 4

    109 D N N

    29$22,0

    00 0 342 SE 10000 S L D N G

    HP 4CYC AUTO F 4

    109 I N N

    30$21,9

    95 0 3 4 NA 2403 S L D N G2.5LI4 AUTO F 4

    109 D N N

    31$21,9

    95 0 365 SE 7732 S M D N G

    HP 4CYC

    A 5SP F 4

    109 D Y N

    32$21,9

    95 0 365 SE 7919 S L D N G

    HP 4CYC

    A 5SP F 4

    109 D Y N

    33$21,9

    95 0 372 SE 11101 S D D N G

    2.5LI4 AUTO F 4

    109 D Y Y

    34 $21,995 1 15 72 XLE 29177 S D L Y G 3.5LV6 A 6SP F 4 109 D N N

    35$21,9

    95 3 3965 SE 32413 S D D N G HP V6

    A 6SP F 4

    109 D Y N

    36$21,9

    95 3 3941 XLE 39480 S L D N G HP V6

    A 6SP F 4

    109 D Y N

    37$21,9

    95 2 2726 XLE 47108 S D L N G

    2.4LI4

    A 5SP F 4

    109 D Y N

    38$21,9

    95 2 2768 NA S D L N G

    3.5LV6

    A 6SP F 4

    109 D N N

    39$21,9

    91 2 2711 XLE 31820 S L L Y G

    3.5LV6

    A 6SP F 4

    109 D N Y

    40

    $21,9

    90 2 27

    1

    3 XLE 30211 S L L N G

    2.4L

    I4 AUTO F 4

    10

    9 D Y Y

    41$21,9

    88 0 332 SE 9522 S D D N G

    2.5LI4 AUTO F 4

    109 D N N

    42$21,8

    78 3 3957 SE 25557 S D D Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    43$21,7

    86 3 3917 XLE 33883 S D D Y G

    3.5LV6

    A 6SP F 4

    109 D Y N

    44$21,5

    95 1 1518 SE 32438 S L D N G

    2.5LI4 AUTO F 4

    109 D Y N

    45$21,5

    14 3 3968 XLE 35822 S D L N G

    3.5LV6

    A 6SP F 4

    109 D Y N

    46$21,5

    00 2 2733 XLE 15802 S D L Y G

    HP 4CYC AUTO F 4

    109 D N Y

    47$21,4

    95 0 365 LE 7919 S D L N G

    HP 4CYC AUTO F 4

    109 D Y Y

    48$20,9

    99 2 2717 XLE 11891 S L D Y G

    2.4LI4 AUTO F 4

    109 D Y N

    49$20,9

    98 3 3942 SE 32395 S D D Y G

    3.5LV6

    A 6SP F 4

    109 D Y N

    50$20,9

    95 0 321 LE 7628 S L L N G

    2.5LI4 AUTO F 4

    109 D Y Y

    11 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    12/45

    51$20,9

    95 0 321 LE 7643 S L L N G

    2.5LI4 AUTO F 4

    109 D Y Y

    52$20,9

    95 0 372 LE 9496 S D L N G

    2.5LI4 AUTO F 4

    109 D Y Y

    53$20,9

    95 2 2741 LE 16886 S D D N G

    HP 4CYC AUTO F 4

    109 D Y Y

    54$20,9

    95 3 3941 SE 23641 S D D N G

    HP 4CYC

    A 5SP F 4

    109 D Y Y

    55

    $20,9

    95 3 39

    2

    6 XLE 27211 S D L N G

    2.4L

    I4

    A 5

    SP F 4

    10

    9 D N N

    56$20,9

    95 3 3972 NA 27998 S D L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    57$20,9

    95 2 2772 XLE 28274 S L L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    58$20,9

    95 1 1532 NA 30258 S D L N G

    2.5LI4 AUTO F 4

    109 D N N

    59$20,9

    95 1 1521 SE 33167 S L D N G

    2.5LI4 AUTO F 4

    109 D N N

    60$20,9

    95 2 2741 SE 33412 S D D N G

    HP 4CYC

    A 5SP F 4

    109 D Y Y

    61$20,9

    91 2 2711 SE 24003 S D L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    62$20,9

    91 3 3924 SE 27459 S D D Y G

    3.5LV6

    A 6SP F 4

    109 D N N

    63$20,9

    91 2 2728 LE 34376 S D L N G

    3.5LV6

    A 6SP F 4

    109 D Y N

    64$20,9

    91 3 3928 XLE 40943 S L L Y G

    3.5LV6

    A 6SP F 4

    109 D Y N

    65$20,9

    90 0 321 LE 3245 S D L N G

    2.5LI4 AUTO F 4

    109 D Y N

    66$20,9

    90 3 3921 XLE 25069 S M L Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    67$20,9

    88 2 2732 XLE 39526 S L L Y G

    2.4LI4 AUTO F 4

    109 D Y Y

    68

    $20,9

    66 2 27

    2

    6 SE 17253 S D D Y G

    2.4L

    I4 AUTO F 4

    10

    9 D Y N

    69$20,7

    86 2 2717 XLE 34389 S L D Y G

    2.4LI4 AUTO F 4

    109 D Y N

    70$20,7

    85 2 27 2 XLE 30601 S D L N G2.4LI4 AUTO F 4

    109 D N Y

    71$20,7

    30 2 2732 NA 24842 S M L N G

    2.4LI4 AUTO F 4

    109 D N N

    72$20,5

    99 2 2720 SE 21904 S L D Y G

    2.4LI4 AUTO F 4

    109 D N N

    73$20,5

    99 2 2718 XLE 27007 S L D Y G

    2.4LI4 AUTO F 4

    109 D N Y

    74$20,2

    88 3 3957 XLE 31512 S D L N G

    2.4LI4 AUTO F 4

    109 D Y Y

    75$19,9

    99 0 317 LE 10270 S D D N G

    2.5LI4 AUTO F 4

    109 D Y N

    76$19,9

    99 1 1517 SE 49311 S L D N G

    2.5LI4 AUTO F 4

    109 D Y N

    77$19,9

    98 0 330 LE 17837 S D D N G

    2.5LI4 AUTO F 4

    109 D N N

    78$19,9

    98 4 5130 XLE 24105 S L L Y G

    2.4LI4 AUTO F 4

    109 D N N

    12 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    13/45

    79$19,9

    98 2 2712 XLE 58457 S M D Y G

    3.5LV6

    A 6SP F 4

    109 D N N

    80$19,9

    95 0 326 LE 4846 S D L N G

    2.5LI4 AUTO F 4

    109 D N N

    81$19,9

    95 0 321 LE 6125 S L L N G

    2.5LI4 AUTO F 4

    109 D Y N

    82$19,9

    95 0 321 LE 7009 S L L N G

    2.5LI4 AUTO F 4

    109 D Y Y

    83

    $19,9

    95 0 3 4 NA 7656 S D L N G

    2.5L

    I4 AUTO F 4

    10

    9 D N Y

    84$19,9

    95 0 326 LE 13884 S L L N G

    2.5LI4 AUTO F 4

    109 D N N

    85$19,9

    95 0 326 LE 14198 S D L N G

    2.5LI4 AUTO F 4

    109 D N N

    86$19,9

    95 2 27 4 NA 15097 S L D Y G2.4LI4 AUTO F 4

    109 D N Y

    87$19,9

    95 2 2717 XLE 28313 S D D N G

    2.4LI4 AUTO F 4

    109 D N N

    88$19,9

    95 0 329 LE 30194 S D D N G

    2.5LI4 AUTO F 4

    109 D Y N

    89$19,9

    95 0 329 LE 34495 S M D N G

    2.5LI4 AUTO F 4

    109 D Y N

    90$19,9

    95 0 329 LE 35591 S D D N G

    2.5LI4 AUTO F 4

    109 D Y N

    91$19,9

    95 2 2717 SE 36663 S L D N G

    2.4LI4 AUTO F 4

    109 D N N

    92$19,9

    95 2 2737 SE 39663 S D D Y G

    3.5LV6

    A 6SP F 4

    109 D Y Y

    93$19,9

    95 2 2726 XLE 49698 S M D Y G

    2.4LI4 AUTO F 4

    109 D Y Y

    94$19,9

    95 1 1530 NA 50431 S L D N G

    2.5LI4 AUTO F 4

    109 D N N

    95$19,9

    93 2 2738 SE 30058 S D D N G

    3.5LV6

    A 6SP F 4

    109 D Y N

    96

    $19,9

    91 0 3 9 LE 1753 S D D N G

    2.5L

    I4 AUTO F 4

    10

    9 D Y N

    97$19,9

    91 0 3 9 LE 2177 S L L N G2.5LI4 AUTO F 4

    109 D Y N

    98$19,9

    91 0 328 LE 2447 S L L N G

    2.5LI4 AUTO F 4

    109 D Y N

    99$19,9

    91 4 51 9 XLE 11507 S L L N G2.4LI4 AUTO F 4

    109 D Y Y

    100$19,9

    91 1 1528 LE 16703 S D L N G

    2.5LI4 AUTO F 4

    109 D Y N

    101$17,9

    95 1 15 9 SE 32010 S D D N G2.5LI4 AUTO F 4

    109 I N Y

    102$15,9

    00 1 15 9 LE 52700 S D D N G2.5LI4 AUTO F 4

    109 I N N

    103$14,9

    95 3 3928 LE 36952 S L L N G

    2.5LI4 AUTO F 4

    109 I N N

    104$13,9

    00 4 51 9 SE 67000 S D D N G2.5LI4 AUTO F 4

    109 I N N

    105$12,4

    95 6 7528 SE 59700 S L D N G

    2.5LI4 AUTO F 4

    109 I N Y

    106$10,9

    00 4 5126 LE 77000 S D D Y G

    2.5LI4 AUTO F 4

    109 I N N

    13 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    14/45

    4) HYPOTHESIS

    In general as the mileage on a car increases, the price of the car gets reduced, also as the

    car gets older the price gets lower. So first hypothesis is

    Hypothesis 1: The price of a Camry sedan is negatively related with its mileage and year

    produced.

    Body styling also plays a major role in the price of a car. Camry is sold in three body types

    i.e. LE, SE, XLE.

    Hypothesis 2(a): The price of a Camry sedan is directly related to the body style LE.

    Hypothesis 2(b): The price of a Camry sedan is directly related to the body style SE.

    Hypothesis 2(c): The price of a Camry sedan is directly related to the body style XLE.

    It is my assumption that price also varies with the exterior and the interior colors of the ca

    too. I have created three categories for exterior colors i.e. dark, medium, and light and two

    interior colors i.e. dark, and light. My hypothesis is:

    Hypothesis 3(a): The price of Camry is directly related to exterior color dark.

    Hypothesis 3(b): The price of Camry is directly related to exterior color medium.

    Hypothesis 3(c): The price of Camry is directly related to exterior color light.

    Hypothesis 3(d): The price of Camry is directly related to interior color dark.

    Hypothesis 3(e): The price of Camry is directly related to interior color light.

    14 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    15/45

    The material used for seating cover also makes a big difference in the price of a car

    Seating covers are made of fabric or leather. So

    Hypothesis 4: The price of a Camry is directly related to the leather seating.

    Engine capacity and style also plays a major role in deciding which car to buy. Many

    people are conscious about the gas price so they need cars which consumes less gas

    whereas some need power so that they can drive faster than other. There are five types of

    engines available in Camry, so my hypothesis will be as follow:

    Hypothesis 5(a): The price of a Camry car is directly related to engine type 2.4 L I4

    Hypothesis 5(b): The price of a Camry car is directly related to engine type 2.5 L I4

    Hypothesis 5(c): The price of a Camry car is directly related to engine type 3.5 L V6

    Hypothesis 5(d): The price of a Camry car is directly related to engine type HP 4

    cylinders.

    Hypothesis 5(e): The price of a Camry car is directly related to engine type HP 6

    cylinders.

    Transmission plays a major role in the speed of the car. It is made of different gea

    systems to cater different segments of the market. In Camry there are three major type of

    transmissions used.

    Hypothesis 6(a): The price of a Camry car is directly related to transmission type

    automatic 4 speed.

    15 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    16/45

    Hypothesis 6(b): The price of a Camry car is directly related to transmission type

    automatic 5 speed.

    Hypothesis 6(c): The price of a Camry car is directly related to transmission type

    automatic 6 speed.

    Most of the car dealers are now selling certified pre owned cars which means that these

    cars were slightly used before but have been sold back to the dealer for an upgraded car

    The dealer has checked these cars and might have made few changes to make them bac

    to the predefined standards. Hence the value of these cars might have been increased.

    Hypothesis 7: The price of a Camry is positively related to its status certified or not

    Sunroof is an extra feature in a car so it will definitely cost more to have it. Therfore

    Hypothesis 8: The price of a Camry car is positively related to the presence of sunroof in

    it.

    Used cars are sold either by private owners or car dealers. If you buy a used car from a

    dealer his price might be higher than from a private owner due to its cost.

    Hypothesis 9: The price of a Camry car is positively related if it sold through dealer.

    5) REGRESSION ANALYSIS AND ITS RESULTS

    The complete regression analysis of the collected data has been attached to the paper as

    an appendix. From this regression, following is the final equation.

    LN (PRICE) = - 104 + 0.0568 YR PRODUCED - 0.0148 LN (MILEAGE) - 0.0723 TYPE_LE+ 0.0884 ENGINE_2.4L I4 + 0.0418 ENGINE_HP 4 CYC+ 0.167 TRANSMISSION_A 6 SPEED + 0.227 OWNER_DEALER+ 0.0007 CERTIFIED_YES + 0.0129 SUNROOF_YES

    Predictor Coef SE Coef T PConstant -104.33 13.58 -7.68 0.000

    16 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    17/45

    YR PRODUCED 0.056802 0.006744 8.42 0.000LN (MILEAGE) -0.014823 0.008992 -1.65 0.103TYPE_LE -0.07225 0.01675 -4.31 0.000ENGINE_2.4L I4 0.08837 0.02100 4.21 0.000ENGINE_HP 4 CYC 0.04178 0.02406 1.74 0.086TRANSMISSION_A 6 SPEED 0.16702 0.01704 9.80 0.000OWNER_DEALER 0.22664 0.03515 6.45 0.000CERTIFIED_YES 0.00073 0.01231 0.06 0.953SUNROOF_YES 0.01291 0.01198 1.08 0.284

    S = 0.0525869 R-Sq = 87.2% R-Sq(adj) = 85.9%

    The equation has proved the following hypothesis.

    Hypothesis 1: The price of a Camry sedan is negatively related with its mileage and yea

    produced.

    Hypothesis 2(a): The price of a Camry sedan is directly related to the body style LE.

    Hypothesis 5(a): The price of a Camry car is directly related to engine type 2.4 L I4

    Hypothesis 5(d): The price of a Camry car is directly related to engine type HP 4 cylinders.

    Hypothesis 6(c): The price of a Camry car is directly related to transmission type

    automatic 6 speed.

    Hypothesis 7: The price of a Camry is positively related to its status certified or not

    Hypothesis 8: The price of a Camry car is positively related to the presence of sunroof in it.

    Hypothesis 9: The price of a Camry car is positively related if it sold through dealer.

    6) CONCLUSION

    Manufacturing systems have undergone great change in recent years and so have the

    accounting systems. These changes include consideration of non-financial measures and

    ABC. There is a need for change in the cost accounting system in order to measure the

    performance of firms more accurately. As the role of support services has increased in

    17 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    18/45

    advanced manufacturing systems (i.e. just-in-time, flexible manufacturing systems and

    computer-integrated manufacturing systems), the need to control overhead costs has

    increased.

    In light of the above results, Toyota should see how they can reduce the co efficient of

    these important factors present in the final regression equation. This reduction will result

    in lowering the final price which might help them in becoming more competitive, and

    hence they can increase their market share as a final result. It will also increase the

    customer awareness, value, and satisfaction.

    Keeping above results in mind they should re evaluate the production process to see if

    they can further increase the efficiency and effectiveness of the system. They should re

    draw the line diagrams so that they are able to make the above mentioned parts

    (parameters which are proven by hypothesis) in faster, cheaper and efficient manne

    without compromising on quality.

    Also they should focus on their ABC system and should re evaluate it and see if they can

    make some modifications in it. They should try to calculate cost of these individual parts

    (parameters which are proven by hypothesis) which might help them find some more

    areas where they can work and may get some reward in term of cost and time reduction

    Cost benefit analysis should be considered in doing these calculations.

    18 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    19/45

    7) REFERENCES

    Horngren, Datar, and Rajan, Cost Accounting - A Managerial Emphasis, 14th Edition,

    Prentice-Hall, 2012.

    J. GREGORY BUSHONG, JOHN C. TALBOTT and DAVID W. CORNELL, Instructiona

    CaseActivity-based Costing Incorporating both Activity and Product Costing

    Accounting Education: an international journal Vol. 17, No. 4, 385403, December 2008

    19 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    20/45

    JAEIL PARK and TIMOTHY W. SIMPSON, Toward an activity-based costing system fo

    product families and product platforms in the early stages of development

    International Journal of Production Research, Vol. 46, No. 1, 1 January 2008, 99130

    K. REZAIE, B. OSTADI and S. A. TORABI, Activity-based costing in flexible

    manufacturing systems with a case study in a forging industry, International Journal oProduction Research,

    Vol. 46, No. 4, 15 February 2008, 10471069

    Pingxin Wang, Fei Du, Dinghua Lei, Thomas W. Lin, The Choice of Cost Drivers in

    Activity-Based Costing: Application at a Chinese Oil Well Cementing Company

    International Journal of Management Vol. 27 No. 2 August 2010.

    John A. Brierley, Christopher J. Cowton, Colin Drury, Product Costing Practices in

    Different Manufacturing Industries: A British Survey, International Journal o

    Management Vol. 24 No. 4 December 2007, 667.

    8) APPENDIX

    A) INITIAL MODEL

    Initially I collected 106 observations but 2 observations were missing the MILEAGE data so they

    were then dropped from the equations. Hence regression analysis was carried out on 104

    observations. Following the first equation that I got from MINITAB 15.

    The regression equation is ---------------------(1)

    20 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    21/45

    PRICE = - 1883530 + 946 YR PRODUCED + 1.17 DISTANCE - 0.0545 MILEAGE- 1519 TYPE_LE - 368 TYPE_SE - 301 TYPE_OTHERS + 592 EXT CLR_DARK+ 709 EXT CLR_LIGHT + 419 INT CLR_DARK - 342 LEATHER SEATS_YES+ 2245 ENGINE_2.4L I4 + 532 ENGINE_2.5L I4 + 1115 ENGINE_HP 4 CYC+ 447 ENGINE_HP V6 + 869 TRANSMISSION_A 5 SPEED+ 4531 TRANSMISSION_A 6 SPEED + 2131 OWNER_DEALER - 385 CERTIFIED_YES+ 473 SUNROOF_YES

    Predictor Coef SE Coef T PConstant -1883530 360143 -5.23 0.000

    YR PRODUCED 946.3 179.2 5.28 0.000DISTANCE 1.172 9.305 0.13 0.900MILEAGE -0.05450 0.01222 -4.46 0.000TYPE_LE -1518.6 509.3 -2.98 0.004TYPE_SE -367.8 475.5 -0.77 0.441TYPE_OTHERS -300.9 537.2 -0.56 0.577EXT CLR_DARK 591.5 558.4 1.06 0.292EXT CLR_LIGHT 708.5 564.1 1.26 0.213INT CLR_DARK 419.1 346.7 1.21 0.230LEATHER SEATS_YES -341.6 410.1 -0.83 0.407ENGINE_2.4L I4 2245.1 793.2 2.83 0.006ENGINE_2.5L I4 532.1 721.1 0.74 0.463ENGINE_HP 4 CYC 1115.2 960.3 1.16 0.249ENGINE_HP V6 447.1 983.5 0.45 0.651TRANSMISSION_A 5 SPEED 868.8 778.1 1.12 0.267

    TRANSMISSION_A 6 SPEED 4531.0 672.6 6.74 0.000OWNER_DEALER 2131.5 951.7 2.24 0.028CERTIFIED_YES -385.2 326.6 -1.18 0.242SUNROOF_YES 473.0 327.4 1.44 0.152

    S = 1381.87 R-Sq = 81.4% R-Sq(adj) = 77.1%

    80706050403020100

    $30,000.00

    $25,000.00

    $20,000.00

    $15,000.00

    $10,000.00

    DISTA NCE

    PIC

    Scatterplot of PR ICE vs DISTANCE

    21 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    22/45

    80000700006000050000400003000020000100000

    $30,000.00

    $25,000.00

    $20,000.00

    $15,000.00

    $10,000.00

    MILEAGE

    PIC

    Sc a t t e r p l o t of P R I C E v s M I L EA G E

    2011201020092008200720062005

    $30,000.00

    $25,000.00

    $20,000.00

    $15,000.00

    $10,000.00

    YR PRODUCED

    PIC

    Scatterplot of PRI CE vs Y R PR ODUCED

    OVERALL FIT

    Hypotheses: H0: 1 = 2 = 3----------------------- = 19 = 0.0

    Ha: At least one of the coefficients is not equal to 0.

    Decision Rule: Reject H0 ifF> 1.65 F(0.05; 19, 84) 1.65

    Do not reject H0 ifF 1.65

    Test Statistic: F= 19.29

    OR

    22 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    23/45

    Decision Rule: Reject H0 ifp value < 0.05

    Do not reject H0 ifp value 0.05

    Test Statistic: p value = 0.000

    Decision: Reject H0

    Conclusion: At least one of the coefficients is not equal to 0. At least one of the 19explanatory variables is important in explaining the variation.

    B) REDUCED MODEL & INDIVIDUAL COEFFICIENT TEST

    Many of the variables have very high p values and low t values so I will test them (only

    those

    whose p values are high) one by one to check whether they are useful in explaining the

    variations. Everything is tested at 5% significant level.

    DISTANCE:

    Hypotheses: H0: 2 = 0.0

    Ha: 2 0.0

    Decision Rule: Reject H0 ift> 1.99 or t< -1.99 t(0.025, 84) 1.99

    Do not reject H0 if -1.99 t 1.99

    Test Statistic: t= 0.13

    OR

    Decision Rule: Reject H0 ifp value < 0.05

    Do not reject H0 ifp value 0.05

    Test Statistic: p value = 0.900

    Decision: Do not reject H0

    DISTANCE is not useful in explaining the variation.

    So the revised eq is ---------------------------------------- (2)

    PRICE = - 1885014 + 947 YR PRODUCED - 0.0544 MILEAGE - 1517 TYPE_LE- 356 TYPE_SE - 295 TYPE_OTHERS + 597 EXT CLR_DARK + 706 EXT CLR_LIGHT+ 415 INT CLR_DARK - 342 LEATHER SEATS_YES + 2252 ENGINE_2.4L I4+ 540 ENGINE_2.5L I4 + 1151 ENGINE_HP 4 CYC + 464 ENGINE_HP V6+ 866 TRANSMISSION_A 5 SPEED + 4542 TRANSMISSION_A 6 SPEED+ 2154 OWNER_DEALER - 380 CERTIFIED_YES + 476 SUNROOF_YES

    23 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    24/45

    Predictor Coef SE Coef T PConstant -1885014 357861 -5.27 0.000YR PRODUCED 947.0 178.1 5.32 0.000MILEAGE -0.05436 0.01209 -4.50 0.000TYPE_LE -1517.2 506.3 -3.00 0.004TYPE_SE -356.4 464.0 -0.77 0.445TYPE_OTHERS -295.3 532.3 -0.55 0.580EXT CLR_DARK 596.5 553.7 1.08 0.284

    EXT CLR_LIGHT 706.0 560.5 1.26 0.211INT CLR_DARK 415.1 343.3 1.21 0.230LEATHER SEATS_YES -342.0 407.7 -0.84 0.404ENGINE_2.4L I4 2252.2 786.6 2.86 0.005ENGINE_2.5L I4 539.6 714.5 0.76 0.452ENGINE_HP 4 CYC 1150.7 912.7 1.26 0.211ENGINE_HP V6 464.4 968.1 0.48 0.633TRANSMISSION_A 5 SPEED 866.4 773.3 1.12 0.266TRANSMISSION_A 6 SPEED 4541.9 663.0 6.85 0.000OWNER_DEALER 2154.3 928.8 2.32 0.023CERTIFIED_YES -379.9 322.0 -1.18 0.241SUNROOF_YES 475.7 324.8 1.46 0.147

    S = 1373.85 R-Sq = 81.4% R-Sq(adj) = 77.4%

    Analysis of Variance

    Source DF SS MS F PRegression 18 699895982 38883110 20.60 0.000Residual Error 85 160433286 1887450Total 103 860329268

    Source DF Seq SSYR PRODUCED 1 114537841MILEAGE 1 127182405TYPE_LE 1 168400020TYPE_SE 1 17199253

    TYPE_OTHERS 1 8584056EXT CLR_DARK 1 13530694EXT CLR_LIGHT 1 56794INT CLR_DARK 1 676046LEATHER SEATS_YES 1 14643563ENGINE_2.4L I4 1 12668769ENGINE_2.5L I4 1 24378673ENGINE_HP 4 CYC 1 74242005ENGINE_HP V6 1 2504978TRANSMISSION_A 5 SPEED 1 4924694TRANSMISSION_A 6 SPEED 1 100593016OWNER_DEALER 1 10024137CERTIFIED_YES 1 1700345SUNROOF_YES 1 4048693

    Unusual Observations

    YRObs PRODUCED PRICE Fit SE Fit Residual St Resid1 2011 29315 26981 731 2334 2.01R2 2009 28995 24765 599 4230 3.42R3 2011 28991 25717 550 3274 2.60R7 2010 26900 24219 670 2681 2.24R13 2011 23999 26590 708 -2591 -2.20R

    24 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    25/45

    81 2011 19995 22525 575 -2530 -2.03R93 2009 19993 22914 487 -2921 -2.27R103 2005 12495 11796 1095 699 0.84 X

    R denotes an observation with a large standardized residual.X denotes an observation whose X value gives it large leverage.

    TYPE SE:

    Hypotheses: H0: 4 = 0.0 (refer to eq 2)

    Ha: 4 0.0

    Decision Rule: Reject H0 ift> 1.99 or t< -1.99 t(0.025, 83) 1.99

    Do not reject H0 if -1.99 t 1.99

    Test Statistic: t= -0.77

    OR

    Decision Rule: Reject H0 ifp value < 0.05

    Do not reject H0 ifp value 0.05

    Test Statistic: p value = 0.445

    Decision: Do not reject H0

    TYPE SE is also not useful in explaining the variation. So the revised eq is------------------------(3)

    PRICE = - 1871467 + 940 YR PRODUCED - 0.0532 MILEAGE - 1315 TYPE_LE- 150 TYPE_OTHERS + 525 EXT CLR_DARK + 670 EXT CLR_LIGHT+ 280 INT CLR_DARK - 302 LEATHER SEATS_YES + 2372 ENGINE_2.4L I4+ 574 ENGINE_2.5L I4 + 1204 ENGINE_HP 4 CYC + 485 ENGINE_HP V6+ 810 TRANSMISSION_A 5 SPEED + 4607 TRANSMISSION_A 6 SPEED+ 2294 OWNER_DEALER - 386 CERTIFIED_YES + 455 SUNROOF_YES

    Predictor Coef SE Coef T PConstant -1871467 356573 -5.25 0.000YR PRODUCED 940.2 177.4 5.30 0.000MILEAGE -0.05323 0.01197 -4.45 0.000TYPE_LE -1314.8 431.3 -3.05 0.003TYPE_OTHERS -150.1 496.4 -0.30 0.763EXT CLR_DARK 525.1 544.5 0.96 0.338EXT CLR_LIGHT 670.0 557.2 1.20 0.232INT CLR_DARK 280.5 294.4 0.95 0.343LEATHER SEATS_YES -301.8 403.4 -0.75 0.456ENGINE_2.4L I4 2371.6 769.2 3.08 0.003ENGINE_2.5L I4 573.8 711.4 0.81 0.422ENGINE_HP 4 CYC 1203.9 907.9 1.33 0.188ENGINE_HP V6 485.5 965.4 0.50 0.616TRANSMISSION_A 5 SPEED 810.2 768.0 1.05 0.294TRANSMISSION_A 6 SPEED 4607.1 656.0 7.02 0.000OWNER_DEALER 2294.2 908.6 2.52 0.013

    25 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    26/45

    CERTIFIED_YES -385.6 321.2 -1.20 0.233SUNROOF_YES 454.8 322.9 1.41 0.163

    S = 1370.57 R-Sq = 81.2% R-Sq(adj) = 77.5%

    Analysis of Variance

    Source DF SS MS F PRegression 17 698782331 41104843 21.88 0.000

    Residual Error 86 161546937 1878453Total 103 860329268

    Source DF Seq SSYR PRODUCED 1 114537841MILEAGE 1 127182405TYPE_LE 1 168400020TYPE_OTHERS 1 2163550EXT CLR_DARK 1 7686805EXT CLR_LIGHT 1 108756INT CLR_DARK 1 4754494LEATHER SEATS_YES 1 21298304ENGINE_2.4L I4 1 6003986ENGINE_2.5L I4 1 22473738

    ENGINE_HP 4 CYC 1 91328610ENGINE_HP V6 1 3020150TRANSMISSION_A 5 SPEED 1 4506246TRANSMISSION_A 6 SPEED 1 108234234OWNER_DEALER 1 11564836CERTIFIED_YES 1 1791345SUNROOF_YES 1 3727011

    Unusual Observations

    YRObs PRODUCED PRICE Fit SE Fit Residual St Resid2 2009 28995 24895 573 4100 3.29R3 2011 28991 25949 458 3042 2.35R

    7 2010 26900 24217 669 2683 2.24R13 2011 23999 26516 700 -2517 -2.14R18 2010 23588 20902 476 2686 2.09R90 2009 19995 22672 427 -2677 -2.06R93 2009 19993 23031 461 -3038 -2.35R103 2005 12495 11815 1093 680 0.82 X

    R denotes an observation with a large standardized residual.X denotes an observation whose X value gives it large leverage.

    REMOVING TYPE OTHERS, EXT CLR DARK, ENGINE HP V6, TRANSMISSIONA 5 SPEED AND COMPARING THE REDUCED MODEL (below) TO THE FULL MODEL(eq 3).

    The regression equation is ------------------ (4)

    PRICE = - 1781671 + 896 YR PRODUCED - 0.0535 MILEAGE - 1381 TYPE_LE+ 189 EXT CLR_LIGHT + 317 INT CLR_DARK - 421 LEATHER SEATS_YES+ 2283 ENGINE_2.4L I4 + 527 ENGINE_2.5L I4 + 1429 ENGINE_HP 4 CYC+ 4536 TRANSMISSION_A 6 SPEED + 2480 OWNER_DEALER - 314 CERTIFIED_YES+ 453 SUNROOF_YES

    26 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    27/45

    Predictor Coef SE Coef T PConstant -1781671 333989 -5.33 0.000YR PRODUCED 895.7 166.3 5.39 0.000MILEAGE -0.05354 0.01168 -4.59 0.000TYPE_LE -1380.9 394.6 -3.50 0.001EXT CLR_LIGHT 188.7 291.8 0.65 0.520INT CLR_DARK 316.6 285.5 1.11 0.270LEATHER SEATS_YES -420.5 366.4 -1.15 0.254ENGINE_2.4L I4 2282.8 738.5 3.09 0.003ENGINE_2.5L I4 527.5 690.6 0.76 0.447

    ENGINE_HP 4 CYC 1429.0 797.2 1.79 0.076TRANSMISSION_A 6 SPEED 4536.0 640.3 7.08 0.000OWNER_DEALER 2480.1 826.6 3.00 0.003CERTIFIED_YES -314.0 298.6 -1.05 0.296SUNROOF_YES 452.7 315.5 1.44 0.155

    S = 1360.25 R-Sq = 80.6% R-Sq(adj) = 77.8%

    Analysis of Variance

    Source DF SS MS F PRegression 13 693805175 53369629 28.84 0.000Residual Error 90 166524092 1850268Total 103 860329268

    Source DF Seq SSYR PRODUCED 1 114537841MILEAGE 1 127182405TYPE_LE 1 168400020EXT CLR_LIGHT 1 6131429INT CLR_DARK 1 4181534LEATHER SEATS_YES 1 22605209ENGINE_2.4L I4 1 6946534ENGINE_2.5L I4 1 20362897ENGINE_HP 4 CYC 1 91406912TRANSMISSION_A 6 SPEED 1 110972667OWNER_DEALER 1 15983480CERTIFIED_YES 1 1283838SUNROOF_YES 1 3810409

    Unusual Observations

    YRObs PRODUCED PRICE Fit SE Fit Residual St Resid1 2011 29315 26895 678 2420 2.05R2 2009 28995 24971 515 4024 3.20R3 2011 28991 25806 435 3185 2.47R7 2010 26900 24058 635 2842 2.36R13 2011 23999 26491 681 -2492 -2.12R18 2010 23588 20984 432 2604 2.02R81 2011 19995 22559 511 -2564 -2.03R90 2009 19995 22644 394 -2649 -2.03R93 2009 19993 23126 403 -3133 -2.41R

    103 2005 12495 11287 959 1208 1.25 X

    R denotes an observation with a large standardized residual.X denotes an observation whose X value gives it large leverage.

    Hypotheses: H0: 4 = 5 = 12= 13 = 0.0

    Ha: At least one of the coefficients is not equal to 0.

    27 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    28/45

    Decision Rule: Reject H0 ifF> 2.48 F(0.05; 4, 86) 2.48

    Do not reject H0 ifF 2.48

    Test Statistic: F= (SSEr SSEf)/4

    SSEf/86

    F= 0.66

    Decision: Do not reject H0.

    Conclusion: Reduced model is better in explaining the variation.

    REMOVING EXT CLR LIGHT, INT CLR DARK, LEATHER SEAT, ANDCOMPARING THE REDUCED MODEL (below) TO THE FULL MODEL (eq 4).

    The regression equation is ------------------(5)

    PRICE = - 1762509 + 886 YR PRODUCED - 0.0540 MILEAGE - 1404 TYPE_LE

    + 2268 ENGINE_2.4L I4 + 690 ENGINE_2.5L I4 + 1622 ENGINE_HP 4 CYC+ 4349 TRANSMISSION_A 6 SPEED + 2558 OWNER_DEALER - 336 CERTIFIED_YES+ 285 SUNROOF_YES

    Predictor Coef SE Coef T P VIFConstant -1762509 328492 -5.37 0.000YR PRODUCED 886.2 163.5 5.42 0.000 2.551MILEAGE -0.05395 0.01156 -4.67 0.000 1.776TYPE_LE -1403.8 384.9 -3.65 0.000 1.605ENGINE_2.4L I4 2267.9 731.7 3.10 0.003 4.507ENGINE_2.5L I4 689.7 671.9 1.03 0.307 5.513ENGINE_HP 4 CYC 1621.8 775.7 2.09 0.039 3.711TRANSMISSION_A 6 SPEED 4349.0 629.0 6.91 0.000 5.410OWNER_DEALER 2557.9 791.0 3.23 0.002 2.215

    CERTIFIED_YES -336.0 297.7 -1.13 0.262 1.227

    S = 1358.21 R-Sq = 80.1% R-Sq(adj) = 77.9%

    Analysis of Variance

    Source DF SS MS F PRegression 10 688770125 68877013 37.34 0.000Residual Error 93 171559142 1844722Total 103 860329268

    Source DF Seq SS

    YR PRODUCED 1 114537841MILEAGE 1 127182405TYPE_LE 1 168400020ENGINE_2.4L I4 1 11711035ENGINE_2.5L I4 1 34429553ENGINE_HP 4 CYC 1 104336628TRANSMISSION_A 6 SPEED 1 107556076OWNER_DEALER 1 17137029CERTIFIED_YES 1 1717365SUNROOF_YES 1 1762172

    28 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    29/45

    Unusual Observations

    YRObs PRODUCED PRICE Fit SE Fit Residual St Resid2 2009 28995 24440 392 4555 3.50R3 2011 28991 25991 403 3000 2.31R7 2010 26900 23744 539 3156 2.53R13 2011 23999 26397 639 -2398 -2.00R18 2010 23588 20957 410 2631 2.03R

    81 2011 19995 22695 453 -2700 -2.11R93 2009 19993 22752 284 -2759 -2.08R103 2005 12495 11322 933 1173 1.19 X

    R denotes an observation with a large standardized residual.X denotes an observation whose X value gives it large leverage.

    Hypotheses: H0: 4 = 5 = 6 = 0.0 (refer to eq 4)

    Ha: At least one of the coefficients is not equal to 0.

    Decision Rule: Reject H0 ifF> 2.72 F(0.05; 3, 90) 2.72

    Do not reject H0 ifF 2.72

    Test Statistic: F= (SSEr SSEf)/3

    SSEf/90

    F= 0.9

    Decision: Do not reject H0.

    Conclusion: Reduced model is better in explaining the variation. Hence eq 5 will be carried

    forward for all further analysis.

    Note: I have not removed two variables certified yes and sunroof yes due to the

    reason that I thought that they might be important in explaining the price

    although they have high p value and if I would tested them they would been

    eliminated. I wanted to check if they do have correlation issues.

    C) MULTICOLLINEARITY:

    Eq 5 indicates that:

    No variable has a VIF greater than 10.

    The average VIF is more than 1

    Two variables has low t value and high p values.

    Four variables have more VIF than 1/1-R2

    29 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    30/45

  • 8/2/2019 Project Paper in Word 97-2003

    31/45

    OWNER_DEALER -0.595 0.2260.000 0.021

    CERTIFIED_YES -0.081 0.063 0.3080.416 0.523 0.001

    SUNROOF_YES 0.075 0.052 0.0440.446 0.598 0.654

    CERTIFIED_YES

    SUNROOF_YES 0.1790.069

    Cell Contents: Pearson correlationP-Value

    As per the above data only one pair is having a more correlation (0.638) than the highestcorrelation of the dependent variable with any other explanatory variables (0.568). Inorder to rectify the issue I removed the variable ENGINE 2.5L I4 and below is the revisedeq along with its VIF and correlations values.

    The regression equation is -------------------------- (6)

    PRICE = - 1895715 + 953 YR PRODUCED - 0.0534 MILEAGE - 1377 TYPE_LE+ 1731 ENGINE_2.4L I4 + 1064 ENGINE_HP 4 CYC+ 3856 TRANSMISSION_A 6 SPEED + 2635 OWNER_DEALER - 378 CERTIFIED_YES+ 289 SUNROOF_YES

    Predictor Coef SE Coef T P VIFConstant -1895715 301863 -6.28 0.000YR PRODUCED 952.7 150.2 6.34 0.000 2.150MILEAGE -0.05341 0.01155 -4.62 0.000 1.772TYPE_LE -1377.4 384.1 -3.59 0.001 1.598ENGINE_2.4L I4 1731.2 512.0 3.38 0.001 2.206ENGINE_HP 4 CYC 1063.8 553.6 1.92 0.058 1.889TRANSMISSION_A 6 SPEED 3856.0 406.4 9.49 0.000 2.257

    OWNER_DEALER 2635.1 787.7 3.35 0.001 2.195CERTIFIED_YES -378.4 294.9 -1.28 0.203 1.203SUNROOF_YES 289.1 291.8 0.99 0.324 1.113

    S = 1358.59 R-Sq = 79.8% R-Sq(adj) = 77.9%

    Analysis of Variance

    Source DF SS MS F PRegression 9 686826156 76314017 41.35 0.000Residual Error 94 173503112 1845778Total 103 860329268

    Source DF Seq SS

    YR PRODUCED 1 114537841MILEAGE 1 127182405TYPE_LE 1 168400020ENGINE_2.4L I4 1 11711035ENGINE_HP 4 CYC 1 50318387TRANSMISSION_A 6 SPEED 1 192601867OWNER_DEALER 1 17949319CERTIFIED_YES 1 2314331SUNROOF_YES 1 1810950

    31 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    32/45

    Unusual Observations

    YRObs PRODUCED PRICE Fit SE Fit Residual St Resid1 2011 29315 26471 407 2844 2.19R2 2009 28995 24514 386 4481 3.44R3 2011 28991 26162 368 2829 2.16R7 2010 26900 23920 511 2980 2.37R18 2010 23588 20862 399 2726 2.10R

    28 2011 22000 20725 734 1275 1.12 X81 2011 19995 22711 452 -2716 -2.12R93 2009 19993 22798 280 -2805 -2.11R103 2005 12495 11580 899 915 0.90 X

    R denotes an observation with a large standardized residual.X denotes an observation whose X value gives it large leverage.

    Correlations: PRICE, YR PRODUCED, MILEAGE, TYPE_LE, ENGINE_2.4L , ...

    PRICE YR PRODUCED MILEAGEYR PRODUCED 0.365

    0.000

    MILEAGE -0.525 -0.5810.000 0.000

    TYPE_LE -0.277 0.426 -0.1770.004 0.000 0.072

    ENGINE_2.4L I4 -0.126 -0.319 0.1190.201 0.001 0.229

    ENGINE_HP 4 CYC -0.346 -0.059 0.1460.000 0.554 0.138

    TRANSMISSION_A 6 0.568 -0.164 0.0270.000 0.097 0.789

    OWNER_DEALER 0.567 0.253 -0.4190.000 0.010 0.000

    CERTIFIED_YES 0.128 0.163 -0.2010.195 0.098 0.040

    SUNROOF_YES 0.167 -0.006 -0.1360.090 0.955 0.169

    TYPE_LE ENGINE_2.4L I4 ENGINE_HP 4 CYCENGINE_2.4L I4 -0.280

    0.004

    ENGINE_HP 4 CYC 0.108 -0.1790.276 0.070

    TRANSMISSION_A 6 -0.319 -0.397 -0.3170.001 0.000 0.001

    OWNER_DEALER -0.104 0.127 -0.5950.296 0.199 0.000

    CERTIFIED_YES 0.119 -0.089 -0.081

    32 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    33/45

    0.230 0.367 0.416

    SUNROOF_YES -0.131 0.055 0.0750.187 0.582 0.446

    TRANSMISSION_A 6 OWNER_DEALER CERTIFIED_YESOWNER_DEALER 0.226

    0.021

    CERTIFIED_YES 0.063 0.308

    0.523 0.001

    SUNROOF_YES 0.052 0.044 0.1790.598 0.654 0.069

    ZCell Contents: Pearson correlation

    P-Value

    Following are the important findings when we compare eq 6 with eq 5.

    No variable has more VIF than 1/1-R2

    The VIF of each variable has reduced.

    No explanatory pair has higher correlation than the highest correlation of dependentvariable.

    Although still there might be some multi collinearity problems present in the revised eq

    but as this equation will be used for forecasting only so I will take this equation further

    from here.

    D) ASSESSING THE ASSUMPTIONS OF THE REGRESSION MODEL

    There are two quantitative variables in the eq 5, so in order to check all four assumptions

    of these variables we need to review the residual plots versus each explanatory variable

    and versus predicted or fitted values. These plots are as under:

    33 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    34/45

    2011201020092008200720062005

    4

    3

    2

    1

    0

    -1

    -2

    Y R PR ODUCED

    StandardizedResdual

    R e s i d ua l s V e r s us Y R P R O D U C E D(response is PRICE)

    2011201020092008200720062005

    5000

    4000

    3000

    2000

    1000

    0

    -1000

    -2000

    -3000

    Y R PRODUCED

    Resdual

    R e s i d ua l s V e r s us Y R P R O D U C E D(response is PRICE)

    34 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    35/45

    80000700006000050000400003000020000100000

    4

    3

    2

    1

    0

    -1

    -2

    MILEAGE

    StandardizedResdual

    Re s i d ua l s V e r s u s M I L EAGE(response is PRICE)

    80000700006000050000400003000020000100000

    5000

    4000

    3000

    2000

    1000

    0

    -1000

    -2000

    -3000

    MILEA GE

    Resdual

    Re s i d ua l s V e r s u s M I L EAGE(response is PRICE)

    35 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    36/45

    28000260002400022000200001800016000140001200010000

    4

    3

    2

    1

    0

    -1

    -2

    Fi t te d Va lue

    StandardizedResdual

    Ve r s u s F i t s(response is PRICE)

    28000260002400022000200001800016000140001200010000

    5000

    4000

    3000

    2000

    1000

    0

    -1000

    -2000

    -3000

    Fi t te d Va lue

    Resdual

    Ve r s u s F i t s(response is PRICE)

    From these plots my findings are:

    There is no pattern between the yr produced and sres or res

    There is some sort of disturbance in mileage vs sres or res which can also be seen

    36 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    37/45

    in res or sres vs fit. In order to further check this I will run checks for all four

    assumptions on mileage to see which type assumption has been violated and wha

    can be done to rectify it.

    1)Assessing the assumption that the relationship is linear:

    As there is no replication so I only conducted DATA SUBSETTING TEST to check whetherthe linearity assumption has been violated.

    The output is as below:

    The regression equation isPRICE = - 1895715 + 953 YR PRODUCED - 0.0534 MILEAGE - 1377 TYPE_LE

    + 1731 ENGINE_2.4L I4 + 1064 ENGINE_HP 4 CYC+ 3856 TRANSMISSION_A 6 SPEED + 2635 OWNER_DEALER - 378 CERTIFIED_YES+ 289 SUNROOF_YES

    Predictor Coef SE Coef T PConstant -1895715 301863 -6.28 0.000YR PRODUCED 952.7 150.2 6.34 0.000MILEAGE -0.05341 0.01155 -4.62 0.000TYPE_LE -1377.4 384.1 -3.59 0.001ENGINE_2.4L I4 1731.2 512.0 3.38 0.001ENGINE_HP 4 CYC 1063.8 553.6 1.92 0.058TRANSMISSION_A 6 SPEED 3856.0 406.4 9.49 0.000OWNER_DEALER 2635.1 787.7 3.35 0.001CERTIFIED_YES -378.4 294.9 -1.28 0.203SUNROOF_YES 289.1 291.8 0.99 0.324

    S = 1358.59 R-Sq = 79.8% R-Sq(adj) = 77.9%

    Analysis of Variance

    Source DF SS MS F PRegression 9 686826156 76314017 41.35 0.000Residual Error 94 173503112 1845778Total 103 860329268

    Lack of fit testPossible interaction in variable MILEAGE (P-Value = 0.018 )

    Possible interaction in variable TRANSMIS (P-Value = 0.000 )

    Overall lack of fit test is significant at P = 0.000

    DATA SUBSETTING WITH 5% LEVEL OF SIGNIFICANCE

    Hypotheses: H0: the relationship is linear

    Ha: the relationship is not linear

    Decision Rule: Reject H0 ifp value < 0.05

    Do not reject H0p value 0.05

    Test Statistic: p-value = 0.018 which is < 0.05

    37 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    38/45

    Decision: Based on sample data it is evidence that there is possible curvature in the

    equation.

    CORRECTION FOR LINEARITY ASSUMPTIONS:

    POLYNOMIAL REGRESSION: I squared the mileage and then run the regression but it

    not good and got the following results

    S = 1375.56 R-Sq = 79.3% R-Sq(adj) = 77.3%

    Which is a clear indication that polynomial will not work. Also I tried the quadratic term

    whose results were worse than the above.

    RECIPROCAL TRANSFORMATION OF EXPLANATORY VARIABLE MILEAGE: I took

    the inverse of mileage and got the following result which also does not show any

    improvement.

    S = 1528.70 R-Sq = 73.7% R-Sq(adj) = 72.0%

    LOG TRANSFORMATION OF MILEAGE: I did the ln and log 10 transformation of themileage but got the following results which did not improved the results.

    (natural log) S = 1402.58 R-Sq = 78.5% R-Sq(adj) = 76.4%(log 10) S = 1411.75 R-Sq = 78.2% R-Sq(adj) = 76.1%

    LOG (NATURAL) TRANSFORMATION OF MILEAGE and PRICE: After taking LN on both

    I got the following regression which is a bit better than the rest, so the following eq is best.

    Also now the p value of the curvature (LN MILE) is 0.06 which is above the critical value.

    The regression equation is -------------------(7)

    LN (PRICE) = - 93.0 + 0.0513 YR PRODUCED - 0.0347 LN (MILEAGE) - 0.0727 TYPE_LE+ 0.0932 ENGINE_2.4L I4 + 0.0704 ENGINE_HP 4 CYC+ 0.180 TRANSMISSION_A 6 SPEED + 0.227 OWNER_DEALER- 0.0176 CERTIFIED_YES + 0.0118 SUNROOF_YES

    Predictor Coef SE Coef T PConstant -93.01 14.20 -6.55 0.000YR PRODUCED 0.051269 0.007050 7.27 0.000LN (MILEAGE) -0.03472 0.01005 -3.45 0.001TYPE_LE -0.07270 0.01824 -3.99 0.000ENGINE_2.4L I4 0.09322 0.02439 3.82 0.000ENGINE_HP 4 CYC 0.07041 0.02623 2.68 0.009TRANSMISSION_A 6 SPEED 0.18026 0.01937 9.31 0.000OWNER_DEALER 0.22697 0.03606 6.29 0.000

    CERTIFIED_YES -0.01762 0.01403 -1.26 0.212SUNROOF_YES 0.01179 0.01375 0.86 0.394

    S = 0.0645036 R-Sq = 81.6% R-Sq(adj) = 79.8%

    Analysis of Variance

    Source DF SS MS F P

    38 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    39/45

    Regression 9 1.73482 0.19276 46.33 0.000Residual Error 94 0.39111 0.00416Total 103 2.12593

    Source DF Seq SSYR PRODUCED 1 0.31959LN (MILEAGE) 1 0.07192TYPE_LE 1 0.51397ENGINE_2.4L I4 1 0.00466ENGINE_HP 4 CYC 1 0.17467

    TRANSMISSION_A 6 SPEED 1 0.47970OWNER_DEALER 1 0.16204CERTIFIED_YES 1 0.00521SUNROOF_YES 1 0.00306

    YRObs PRODUCED LN (PRICE) Fit SE Fit Residual St Resid1 2011 9.9030 9.9748 0.0228 -0.0718 -1.192 2011 9.9030 9.9673 0.0212 -0.0643 -1.053 2011 9.9986 10.0542 0.0243 -0.0556 -0.934 2011 9.9030 9.9632 0.0204 -0.0602 -0.985 2011 9.9518 9.9534 0.0185 -0.0016 -0.036 2011 10.2396 10.2074 0.0195 0.0323 0.527 2011 9.9032 9.9571 0.0187 -0.0539 -0.87

    8 2009 10.2749 10.1065 0.0209 0.1684 2.76R9 2011 10.2396 10.1967 0.0183 0.0429 0.6910 2011 9.9032 9.9314 0.0156 -0.0282 -0.4511 2011 9.9032 9.9385 0.0176 -0.0353 -0.5712 2011 9.9520 9.9356 0.0175 0.0165 0.2713 2011 9.9520 9.9355 0.0175 0.0165 0.2714 2011 9.9032 10.0257 0.0215 -0.1225 -2.01R15 2011 9.9986 10.0664 0.0269 -0.0678 -1.1616 2011 10.1655 10.1878 0.0176 -0.0223 -0.3617 2011 9.9986 10.0656 0.0269 -0.0670 -1.1418 2011 9.9756 10.0047 0.0264 -0.0291 -0.4919 2011 10.0858 10.1745 0.0183 -0.0887 -1.4320 2011 10.2747 10.1862 0.0176 0.0886 1.4321 2007 10.0429 9.9857 0.0255 0.0572 0.9722 2011 10.2859 10.1987 0.0194 0.0872 1.42

    23 2011 9.9520 9.9280 0.0175 0.0241 0.3924 2011 9.9983 10.0064 0.0189 -0.0081 -0.1325 2011 10.2033 10.1806 0.0174 0.0227 0.3626 2010 10.1263 10.1283 0.0148 -0.0020 -0.0327 2011 9.9988 9.8481 0.0332 0.1507 2.73R28 2011 9.9034 9.9134 0.0148 -0.0100 -0.1629 2011 9.9986 9.9952 0.0198 0.0033 0.0530 2011 10.1258 10.1627 0.0178 -0.0369 -0.5931 2007 9.9030 9.8821 0.0240 0.0209 0.3532 2009 9.9522 9.9717 0.0194 -0.0195 -0.3233 2010 10.0685 9.9297 0.0183 0.1388 2.24R34 2010 10.0690 10.1244 0.0148 -0.0554 -0.8835 2010 10.1999 10.0621 0.0242 0.1377 2.30R36 2009 10.0429 10.0540 0.0142 -0.0111 -0.1837 2011 9.9032 9.9206 0.0171 -0.0174 -0.28

    38 2011 9.9032 9.9198 0.0172 -0.0166 -0.2739 2009 9.9032 9.9929 0.0191 -0.0896 -1.4540 2009 10.0769 10.0502 0.0138 0.0267 0.4241 2009 9.9758 9.9685 0.0282 0.0073 0.1342 2010 9.9030 9.8453 0.0158 0.0577 0.9243 2009 9.9520 9.8758 0.0285 0.0762 1.3244 2009 9.9507 9.9588 0.0184 -0.0082 -0.1345 2011 9.9034 9.9119 0.0177 -0.0085 -0.1446 2008 10.0336 9.9201 0.0180 0.1135 1.8347 2009 9.9330 9.9681 0.0171 -0.0351 -0.56

    39 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    40/45

    48 2009 10.0856 10.0536 0.0135 0.0320 0.5149 2008 9.9520 9.8856 0.0279 0.0665 1.1450 2009 9.9518 10.0462 0.0137 -0.0943 -1.5051 2007 9.9034 9.8623 0.0207 0.0411 0.6752 2009 9.9393 9.9638 0.0170 -0.0244 -0.3953 2008 10.0560 9.9995 0.0156 0.0565 0.9054 2008 9.9518 9.9934 0.0157 -0.0416 -0.6655 2011 10.1302 10.0803 0.0238 0.0499 0.8356 2008 9.9932 9.9927 0.0157 0.0005 0.0157 2009 9.9330 9.9727 0.0187 -0.0397 -0.6458 2008 9.9520 9.9093 0.0173 0.0427 0.69

    59 2009 10.0429 10.0300 0.0134 0.0128 0.2060 2008 9.9518 9.9961 0.0154 -0.0442 -0.7161 2008 9.9520 9.9896 0.0157 -0.0375 -0.6062 2009 9.9520 10.0405 0.0140 -0.0885 -1.4063 2009 9.9032 9.9592 0.0170 -0.0560 -0.9064 2010 10.1760 10.1090 0.0176 0.0670 1.0865 2010 9.9986 10.0965 0.0153 -0.0979 -1.5666 2009 9.9031 10.0266 0.0135 -0.1234 -1.9667 2011 9.9032 9.8760 0.0187 0.0272 0.4468 2009 9.9983 9.9511 0.0182 0.0472 0.7669 2010 9.9520 9.9150 0.0194 0.0371 0.6070 2009 9.9420 9.9683 0.0189 -0.0263 -0.4371 2009 10.0384 10.0546 0.0166 -0.0162 -0.2672 2008 9.9178 9.8984 0.0183 0.0194 0.3173 2009 9.9984 10.0540 0.0167 -0.0556 -0.89

    74 2010 9.7978 9.7683 0.0298 0.0296 0.5275 2008 9.9522 9.9727 0.0150 -0.0205 -0.3376 2008 9.9986 9.9727 0.0150 0.0259 0.4177 2010 9.9802 9.8949 0.0201 0.0853 1.3978 2010 9.9520 9.9118 0.0198 0.0403 0.6679 2010 10.0813 10.0017 0.0211 0.0795 1.3080 2009 10.0430 10.0405 0.0138 0.0025 0.0481 2009 9.9520 9.9248 0.0263 0.0272 0.4682 2008 9.9890 9.9712 0.0150 0.0179 0.2883 2009 9.9518 9.9492 0.0208 0.0026 0.0484 2009 9.9420 9.9349 0.0186 0.0072 0.1285 2011 9.9032 9.8714 0.0195 0.0319 0.5286 2009 10.0643 10.0385 0.0140 0.0258 0.4187 2011 9.9032 9.8703 0.0197 0.0329 0.5488 2008 9.9765 9.9692 0.0151 0.0072 0.12

    89 2009 9.9032 9.9503 0.0174 -0.0470 -0.7690 2008 9.6155 9.5762 0.0274 0.0392 0.6791 2008 9.9986 9.9658 0.0152 0.0327 0.5292 2009 9.9517 9.9418 0.0190 0.0099 0.1693 2009 9.9032 10.0287 0.0152 -0.1255 -2.00R94 2008 9.9518 9.9646 0.0153 -0.0127 -0.2095 2009 9.9986 9.9239 0.0195 0.0746 1.2196 2009 10.0430 10.0273 0.0152 0.0157 0.2597 2010 9.9034 9.8804 0.0222 0.0230 0.3898 2009 9.9032 9.9339 0.0199 -0.0306 -0.5099 2010 9.9032 9.8972 0.0218 0.0060 0.10100 2010 9.6741 9.6665 0.0296 0.0076 0.13101 2009 9.9034 10.0211 0.0161 -0.1177 -1.88102 2005 9.4331 9.4199 0.0428 0.0132 0.27 X103 2007 9.5396 9.5770 0.0267 -0.0374 -0.64

    104 2007 9.2965 9.4995 0.0290 -0.2030 -3.52R

    R denotes an observation with a large standardized residual.X denotes an observation whose X value gives it large leverage.

    Lack of fit testPossible interaction in variable YR PRODU (P-Value = 0.007 )

    Possible curvature in variable LN (MILE (P-Value = 0.060 )

    40 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    41/45

    Possible interaction in variable LN (MILE (P-Value = 0.009 )

    Possible interaction in variable TRANSMIS (P-Value = 0.006 )

    Possible curvature in variable OWNER_DE (P-Value = 0.036 )

    Overall lack of fit test is significant at P = 0.006

    THE REVISED GRAPHS ARE AS BELOW:

    1110987

    3

    2

    1

    0

    -1

    -2

    -3

    -4

    LN (MILEA GE)

    StandardizedResdual

    R e s i d ua l s V e r s u s L N ( M I L E A G E )(response is LN (PRICE))

    41 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    42/45

    10.310.210.110.09.99.89.79.69.59.4

    3

    2

    1

    0

    -1

    -2

    -3

    -4

    Fi t te d Va lue

    StandardizedResdual

    Ve r s u s F i t s(response is LN (PRICE))

    2011201020092008200720062005

    3

    2

    1

    0

    -1

    -2

    -3

    -4

    Y R PR ODUCED

    StandardizedResdual

    R e s i d ua l s V e r s us Y R P R O D U C E D(response is LN (PRICE))

    2)Assessing the assumption that the variance around the regression is

    constant: as the magnitude of the fitted values (graph on page 23) of the response

    variable get larger, the

    42 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    43/45

    magnitudes of the residual get larger, which is an evidence of non constant variance.

    Hypotheses: H0: the variance is constant (testing at 5% of significance)

    Ha: the variance is not constant

    Decision Rule: Reject H0 if Q > Z (1.645)

    Test Statistic: Q is the Szroeters test

    By using minitab Q comes out to be -0.462893

    Decision: As Q value is less than 1.645 so we cannot accept Ha. Hence the variance is constant.

    3)Assessing the assumption that the disturbances are normally distributed:

    According to empirical rule, approx 68% of the sres should be between -1 and +1,

    approx 95% should be between -2 and-2, and 99.7% should be between -3 and 3.

    From the graph between sres and fitted value on page 25, it is clear that most of the

    values are within +3 to -3 range. So initially it does not appear to violate the assumptionAlso from the data from page 22-24 I calculated that only 6 observations are above +2 or

    -2 which is around 5.7%, hence 94.3% data is within +2, -2 limits which again gives an

    idea that there might not be any violation of this assumption.

    In order to make sure, I conducted the test which is as below:

    Hypotheses: H0: Disturbances are normal (5% level of significance)

    Ha: Disturbances are not normal

    Decision Rule: Reject H0 ifp value < 0.05

    Do not reject H0p value 0.05

    Decision: From minitab using A.D test, p values comes out to be 0.142, which shows

    that disturbances are normal and no violation has been done. The graph is as below.

    43 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    44/45

    43210-1-2-3-4

    99.9

    99

    95

    90

    80

    70

    60

    50

    40

    30

    20

    10

    5

    1

    0.1

    SRES1

    Pc

    Mean 0.004720

    StDev 1.017

    N 104

    AD 0.563

    P-Value 0.142

    Probabi l i t y P l o t o f S RES1Normal

    Further I decided to check the data for some unusual observations with extreme values

    (may be typing errors as well). With minitab I found Cook I, DFIT 1 values (available in the

    minitab file-attached), which I compared with 2 (sqrt (k+1)/n) which came out to be 0.62

    for Dfit and 1 for COOK. I found that five observations were out of 0.62 range i.eobservation # 2, 7, 18, 28 & 51. I removed all these observations and ran the regression

    again, which is below with an improved R and adj R.

    The regression equation is --------------------- (8)

    LN (PRICE) = - 92.1 + 0.0508 YR PRODUCED - 0.0229 LN (MILEAGE) - 0.0708 TYPE_LE+ 0.0887 ENGINE_2.4L I4 + 0.0561 ENGINE_HP 4 CYC+ 0.170 TRANSMISSION_A 6 SPEED + 0.250 OWNER_DEALER- 0.0099 CERTIFIED_YES + 0.0149 SUNROOF_YES

    Predictor Coef SE Coef T PConstant -92.14 13.09 -7.04 0.000

    YR PRODUCED 0.050769 0.006500 7.81 0.000LN (MILEAGE) -0.022899 0.009110 -2.51 0.014TYPE_LE -0.07076 0.01784 -3.97 0.000ENGINE_2.4L I4 0.08872 0.02179 4.07 0.000ENGINE_HP 4 CYC 0.05612 0.02436 2.30 0.024TRANSMISSION_A 6 SPEED 0.16955 0.01775 9.55 0.000OWNER_DEALER 0.25035 0.03539 7.07 0.000CERTIFIED_YES -0.00993 0.01269 -0.78 0.436SUNROOF_YES 0.01492 0.01253 1.19 0.237

    S = 0.0561215 R-Sq = 85.7% R-Sq(adj) = 84.3%

    After removing the observations, I once again run the regression (as shown above eq 8and then re calculated the standard for DFIT which now came to 0.64 and COOK to be 1

    again. I observed that four more observations (1,20,24 & 47) to be out of this range. removed these observations as well and got the following equation.

    The regression equation is --------------------(9)

    LN (PRICE) = - 104 + 0.0568 YR PRODUCED - 0.0148 LN (MILEAGE) - 0.0723 TYPE_LE+ 0.0884 ENGINE_2.4L I4 + 0.0418 ENGINE_HP 4 CYC+ 0.167 TRANSMISSION_A 6 SPEED + 0.227 OWNER_DEALER+ 0.0007 CERTIFIED_YES + 0.0129 SUNROOF_YES

    44 TERMPAPER | ACT 412

  • 8/2/2019 Project Paper in Word 97-2003

    45/45

    Predictor Coef SE Coef T PConstant -104.33 13.58 -7.68 0.000YR PRODUCED 0.056802 0.006744 8.42 0.000LN (MILEAGE) -0.014823 0.008992 -1.65 0.103TYPE_LE -0.07225 0.01675 -4.31 0.000ENGINE_2.4L I4 0.08837 0.02100 4.21 0.000ENGINE_HP 4 CYC 0.04178 0.02406 1.74 0.086TRANSMISSION_A 6 SPEED 0.16702 0.01704 9.80 0.000OWNER_DEALER 0.22664 0.03515 6.45 0.000

    CERTIFIED_YES 0.00073 0.01231 0.06 0.953SUNROOF_YES 0.01291 0.01198 1.08 0.284

    S = 0.0525869 R-Sq = 87.2% R-Sq(adj) = 85.9%

    Analysis of Variance

    Source DF SS MS F PRegression 9 1.60671 0.17852 64.56 0.000Residual Error 85 0.23506 0.00277Total 94 1.84177

    Source DF Seq SSYR PRODUCED 1 0.33014LN (MILEAGE) 1 0.01426TYPE_LE 1 0.55630ENGINE_2.4L I4 1 0.00151ENGINE_HP 4 CYC 1 0.19389TRANSMISSION_A 6 SPEED 1 0.38516OWNER_DEALER 1 0.12202CERTIFIED_YES 1 0.00022SUNROOF_YES 1 0.00321

    Unusual Observations

    YR

    Obs PRODUCED LN (PRICE) Fit SE Fit Residual St Resid1 2011 10.2747 10.1710 0.0154 0.1037 2.06R68 2009 9.9034 10.0146 0.0133 -0.1112 -2.19R72 2011 9.9032 10.0042 0.0181 -0.1010 -2.04R81 2009 9.9032 10.0340 0.0129 -0.1307 -2.56R84 2009 9.9031 10.0252 0.0112 -0.1220 -2.37R94 2005 9.4331 9.4063 0.0356 0.0268 0.69 X95 2007 9.2965 9.4727 0.0248 -0.1762 -3.80R

    R denotes an observation with a large standardized residual.X denotes an observation whose X value gives it large leverage.

    4)Assessing the assumption that the disturbances are independent: This

    assumption is checked when using a time series data, since in my analysis there is notime series so I will not check this assumption.