Fuel Economoy EDA
Uzair Ahmad
2019-12-06
Connceting Data, Information and Knowledge
Typically answering question: - How has the fuel economy changed over time? - Are there any other interesting insights or trends?
EDA
Dataset
The Fuel Economogy dataset
- 42002 Rows
- x Variables
## barrels08 barrelsA08 charge120 charge240
## Min. : 0.06 Min. : 0.0000 Min. :0 Min. : 0.00000
## 1st Qu.:14.33 1st Qu.: 0.0000 1st Qu.:0 1st Qu.: 0.00000
## Median :16.48 Median : 0.0000 Median :0 Median : 0.00000
## Mean :17.36 Mean : 0.2201 Mean :0 Mean : 0.03609
## 3rd Qu.:19.39 3rd Qu.: 0.0000 3rd Qu.:0 3rd Qu.: 0.00000
## Max. :47.09 Max. :18.3117 Max. :0 Max. :12.00000
##
## city08 city08U cityA08 cityA08U
## Min. : 6.00 Min. : 0.000 Min. : 0.0000 Min. : 0.0000
## 1st Qu.: 15.00 1st Qu.: 0.000 1st Qu.: 0.0000 1st Qu.: 0.0000
## Median : 17.00 Median : 0.000 Median : 0.0000 Median : 0.0000
## Mean : 18.21 Mean : 5.495 Mean : 0.6161 Mean : 0.4662
## 3rd Qu.: 20.00 3rd Qu.: 12.274 3rd Qu.: 0.0000 3rd Qu.: 0.0000
## Max. :150.00 Max. :150.000 Max. :145.0000 Max. :145.0835
##
## cityCD cityE cityUF co2
## Min. :0.000000 Min. : 0.0000 Min. :0.000000 Min. : -1.00
## 1st Qu.:0.000000 1st Qu.: 0.0000 1st Qu.:0.000000 1st Qu.: -1.00
## Median :0.000000 Median : 0.0000 Median :0.000000 Median : -1.00
## Mean :0.000471 Mean : 0.2741 Mean :0.001279 Mean : 80.11
## 3rd Qu.:0.000000 3rd Qu.: 0.0000 3rd Qu.:0.000000 3rd Qu.: -1.00
## Max. :5.350000 Max. :122.0000 Max. :0.896000 Max. :847.00
##
## co2A co2TailpipeAGpm co2TailpipeGpm comb08
## Min. : -1.000 Min. : 0.00 Min. : 0.0 Min. : 7.00
## 1st Qu.: -1.000 1st Qu.: 0.00 1st Qu.: 386.4 1st Qu.: 17.00
## Median : -1.000 Median : 0.00 Median : 447.0 Median : 20.00
## Mean : 5.713 Mean : 17.72 Mean : 468.5 Mean : 20.46
## 3rd Qu.: -1.000 3rd Qu.: 0.00 3rd Qu.: 523.0 3rd Qu.: 23.00
## Max. :713.000 Max. :713.00 Max. :1269.6 Max. :136.00
##
## comb08U combA08 combA08U combE
## Min. : 0.000 Min. : 0.0000 Min. : 0.0000 Min. : 0.0000
## 1st Qu.: 0.000 1st Qu.: 0.0000 1st Qu.: 0.0000 1st Qu.: 0.0000
## Median : 0.000 Median : 0.0000 Median : 0.0000 Median : 0.0000
## Mean : 6.149 Mean : 0.6771 Mean : 0.5042 Mean : 0.2804
## 3rd Qu.: 14.273 3rd Qu.: 0.0000 3rd Qu.: 0.0000 3rd Qu.: 0.0000
## Max. :136.000 Max. :133.0000 Max. :133.2662 Max. :121.0000
##
## combinedCD combinedUF cylinders displ
## Min. :0.000000 Min. :0.000000 Min. : 2.000 Min. :0.000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.: 4.000 1st Qu.:2.200
## Median :0.000000 Median :0.000000 Median : 6.000 Median :3.000
## Mean :0.000363 Mean :0.001261 Mean : 5.722 Mean :3.302
## 3rd Qu.:0.000000 3rd Qu.:0.000000 3rd Qu.: 6.000 3rd Qu.:4.300
## Max. :4.800000 Max. :0.888000 Max. :16.000 Max. :8.400
## NA's :171 NA's :169
## drive engId eng_dscr
## Front-Wheel Drive :13939 Min. : 0 :15899
## Rear-Wheel Drive :13539 1st Qu.: 0 (FFS) : 8827
## 4-Wheel or All-Wheel Drive: 6648 Median : 186 SIDI : 4902
## All-Wheel Drive : 2713 Mean : 8377 (FFS) CA model : 926
## 4-Wheel Drive : 1328 3rd Qu.: 4301 (FFS) (MPFI): 734
## : 1189 Max. :69102 FFV : 683
## (Other) : 725 (Other) : 8110
## feScore fuelCost08 fuelCostA08 fuelType
## Min. :-1.0000 Min. : 500 Min. : 0.00 Regular :25997
## 1st Qu.:-1.0000 1st Qu.:1950 1st Qu.: 0.00 Premium :11067
## Median :-1.0000 Median :2350 Median : 0.00 Gasoline or E85: 1287
## Mean : 0.2389 Mean :2378 Mean : 90.06 Diesel : 1142
## 3rd Qu.:-1.0000 3rd Qu.:2700 3rd Qu.: 0.00 Electricity : 168
## Max. :10.0000 Max. :7350 Max. :3800.00 Premium or E85 : 125
## (Other) : 295
## fuelType1 ghgScore ghgScoreA highway08
## Diesel : 1142 Min. :-1.000 Min. :-1.0000 Min. : 9.00
## Electricity : 168 1st Qu.:-1.000 1st Qu.:-1.0000 1st Qu.: 20.00
## Midgrade Gasoline: 100 Median :-1.000 Median :-1.0000 Median : 24.00
## Natural Gas : 60 Mean : 0.237 Mean :-0.9224 Mean : 24.35
## Premium Gasoline :11267 3rd Qu.:-1.000 3rd Qu.:-1.0000 3rd Qu.: 28.00
## Regular Gasoline :27344 Max. :10.000 Max. : 8.0000 Max. :123.00
##
## highway08U highwayA08 highwayA08U highwayCD
## Min. : 0.000 Min. : 0.0000 Min. : 0.0000 Min. :0.000000
## 1st Qu.: 0.000 1st Qu.: 0.0000 1st Qu.: 0.0000 1st Qu.:0.000000
## Median : 0.000 Median : 0.0000 Median : 0.0000 Median :0.000000
## Mean : 7.283 Mean : 0.7826 Mean : 0.5751 Mean :0.000242
## 3rd Qu.: 17.546 3rd Qu.: 0.0000 3rd Qu.: 0.0000 3rd Qu.:0.000000
## Max. :123.340 Max. :121.0000 Max. :121.2005 Max. :4.060000
##
## highwayE highwayUF hlv hpv
## Min. : 0.0000 Min. :0.000000 Min. : 0.00 Min. : 0.00
## 1st Qu.: 0.0000 1st Qu.:0.000000 1st Qu.: 0.00 1st Qu.: 0.00
## Median : 0.0000 Median :0.000000 Median : 0.00 Median : 0.00
## Mean : 0.2885 Mean :0.001237 Mean : 2.02 Mean : 10.36
## 3rd Qu.: 0.0000 3rd Qu.:0.000000 3rd Qu.: 0.00 3rd Qu.: 0.00
## Max. :120.0000 Max. :0.877000 Max. :49.00 Max. :195.00
##
## id lv2 lv4 make
## Min. : 1 Min. : 0.000 Min. : 0.000 Chevrolet: 3944
## 1st Qu.:10021 1st Qu.: 0.000 1st Qu.: 0.000 Ford : 3284
## Median :20042 Median : 0.000 Median : 0.000 Dodge : 2559
## Mean :20154 Mean : 1.815 Mean : 6.139 GMC : 2471
## 3rd Qu.:30331 3rd Qu.: 0.000 3rd Qu.:13.000 Toyota : 2010
## Max. :40434 Max. :41.000 Max. :55.000 BMW : 1856
## (Other) :23957
## model mpgData phevBlended pv2
## F150 Pickup 2WD: 215 N:27367 false:40005 Min. : 0.00
## F150 Pickup 4WD: 193 Y:12714 true : 76 1st Qu.: 0.00
## Mustang : 192 Median : 0.00
## Jetta : 190 Mean : 13.56
## Truck 2WD : 187 3rd Qu.: 0.00
## Camaro : 174 Max. :194.00
## (Other) :38930
## pv4 range rangeCity rangeCityA
## Min. : 0.00 Min. : 0.0000 Min. : 0.0000 Min. : 0.00000
## 1st Qu.: 0.00 1st Qu.: 0.0000 1st Qu.: 0.0000 1st Qu.: 0.00000
## Median : 0.00 Median : 0.0000 Median : 0.0000 Median : 0.00000
## Mean : 33.82 Mean : 0.6164 Mean : 0.5786 Mean : 0.06753
## 3rd Qu.: 91.00 3rd Qu.: 0.0000 3rd Qu.: 0.0000 3rd Qu.: 0.00000
## Max. :192.00 Max. :335.0000 Max. :333.1115 Max. :103.03000
##
## rangeHwy rangeHwyA trany UCity
## Min. : 0.0000 Min. : 0.00000 Automatic 4-spd:11045 Min. : 0.00
## 1st Qu.: 0.0000 1st Qu.: 0.00000 Manual 5-spd : 8351 1st Qu.: 18.11
## Median : 0.0000 Median : 0.00000 Automatic 3-spd: 3151 Median : 21.30
## Mean : 0.5639 Mean : 0.06257 Automatic (S6) : 2984 Mean : 22.98
## 3rd Qu.: 0.0000 3rd Qu.: 0.00000 Manual 6-spd : 2671 3rd Qu.: 25.70
## Max. :346.9000 Max. :90.55000 Automatic 5-spd: 2198 Max. :224.80
## (Other) : 9681
## UCityA UHighway UHighwayA
## Min. : 0.0000 Min. : 0.00 Min. : 0.000
## 1st Qu.: 0.0000 1st Qu.: 27.66 1st Qu.: 0.000
## Median : 0.0000 Median : 33.02 Median : 0.000
## Mean : 0.7894 Mean : 34.11 Mean : 1.077
## 3rd Qu.: 0.0000 3rd Qu.: 38.84 3rd Qu.: 0.000
## Max. :207.2622 Max. :182.70 Max. :173.144
##
## VClass year youSaveSpend guzzler
## Compact Cars : 5751 Min. :1984 Min. :-29000 :37704
## Subcompact Cars : 5036 1st Qu.:1991 1st Qu.: -5750 G: 1398
## Midsize Cars : 4697 Median :2002 Median : -4000 S: 15
## Standard Pickup Trucks : 2354 Mean :2001 Mean : -4135 T: 964
## Sport Utility Vehicle - 4WD: 2090 3rd Qu.:2011 3rd Qu.: -2000
## Large Cars : 2072 Max. :2019 Max. : 5250
## (Other) :18081
## trans_dscr tCharger sCharger atvType
## :25034 Mode:logical :39285 :36707
## CLKUP : 7809 TRUE:6302 S: 796 FFV : 1412
## SIL : 2189 NA's:33779 Diesel : 1070
## 2MODE CLKUP : 1235 Hybrid : 539
## Creeper : 525 EV : 168
## EMS 2MODE CLKUP: 520 Plug-in Hybrid: 107
## (Other) : 2769 (Other) : 78
## fuelType2 rangeA evMotor mfrCode
## :38534 :38539 :39345 :30818
## E85 : 1412 290 : 74 288V Ni-MH : 122 GMX : 1366
## Electricity: 107 270 : 57 245V Ni-MH : 48 BMX : 1072
## Natural Gas: 20 280 : 54 144V Li-Ion: 28 FMX : 792
## Propane : 8 310 : 41 270V Li-Ion: 26 CRX : 703
## 277 : 38 330V Ni-MH : 26 TYX : 662
## (Other): 1278 (Other) : 486 (Other): 4668
## c240Dscr charge240b c240bDscr
## :40016 Min. :0.000000 :40018
## 3.6 kW charger : 4 1st Qu.:0.000000 3.6 kW charger : 2
## 6.6 kW charger : 2 Median :0.000000 6.6 kW charger : 4
## 7.2 kW charger : 2 Mean :0.007497 80 amp dual charger: 54
## single charger : 3 3rd Qu.:0.000000 dual charger : 3
## standard charger: 54 Max. :8.500000
##
## createdOn modifiedOn
## Tue Jan 01 00:00:00 EST 2013:34217 Tue Jan 01 00:00:00 EST 2013:29437
## Fri Jun 06 00:00:00 EDT 2014: 100 Mon Sep 26 00:00:00 EDT 2016: 6346
## Thu Jul 07 00:00:00 EDT 2016: 100 Wed Apr 05 00:00:00 EDT 2017: 911
## Thu Jul 31 00:00:00 EDT 2014: 95 Wed Dec 20 00:00:00 EST 2017: 280
## Wed Jun 13 00:00:00 EDT 2018: 93 Tue Nov 22 00:00:00 EST 2016: 191
## Thu Aug 03 00:00:00 EDT 2017: 86 Fri Dec 02 00:00:00 EST 2016: 183
## (Other) : 5390 (Other) : 2733
## startStop phevCity phevHwy phevComb
## :31704 Min. : 0.0000 Min. : 0.0000 Min. : 0.0000
## N: 5695 1st Qu.: 0.0000 1st Qu.: 0.0000 1st Qu.: 0.0000
## Y: 2682 Median : 0.0000 Median : 0.0000 Median : 0.0000
## Mean : 0.1229 Mean : 0.1234 Mean : 0.1225
## 3rd Qu.: 0.0000 3rd Qu.: 0.0000 3rd Qu.: 0.0000
## Max. :97.0000 Max. :81.0000 Max. :88.0000
## The Dependent Variable
UCity
Definition - UCity - unadjusted city MPG for fuelType1; - UCityA - unadjusted city MPG for fuelType2;
- Remarks
- The median UCity (21.30) seems to be unchanged over the years
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 18.11 21.30 22.98 25.70 224.80
- The right-skewed distribution of UCity tells a tale of a different class of vehicles in last ten years
- There are 169 vehicles having UCity > 75. And as was expected, dominant majority (150+) are Electric vehicles.
- 25 zero UCity vehicles:
## [1] 25## make model fuelType
## 8128 Ford F150 Dual-fuel 2WD (CNG) Gasoline or natural gas
## 8129 Ford F150 Dual-fuel 4WD (CNG) Gasoline or natural gas
## 8130 Ford F150 Dual-fuel 2WD (LPG) Gasoline or propane
## 8131 Ford F150 Dual-fuel 4WD (LPG) Gasoline or propane
## 9175 Dodge Ram Van 2500 2WD CNG CNG
## 9176 Dodge Ram Wagon 2500 2WD CNG CNG
## 9184 Ford F150 Dual-fuel 2WD (CNG) Gasoline or natural gas
## 9185 Ford F150 Dual-fuel 4WD (CNG) Gasoline or natural gas
## 9186 Ford F150 Dual-fuel 2WD (LPG) Gasoline or propane
## 9187 Ford F150 Dual-fuel 4WD (LPG) Gasoline or propane
## 10283 Ford F150 Dual-fuel 2WD (LPG) Gasoline or propane
## 10284 Ford F150 Dual-fuel 4WD (LPG) Gasoline or propane
## 11585 Ford F150 Dual-fuel 2WD (LPG) Gasoline or propane
## 11586 Ford F150 Dual-fuel 4WD (LPG) Gasoline or propane
## 11587 Chevrolet Express Cargo (Bi-fuel) Gasoline or natural gas
## 11588 Chevrolet Express Passenger (Bi-fuel) Gasoline or natural gas
## 11589 GMC Savana (cargo) (Bi-fuel) Gasoline or natural gas
## 11590 GMC Savana Passenger (Bi-fuel) Gasoline or natural gas
## 11592 Chevrolet Express Cargo (dedicated CNG) CNG
## 11593 Chevrolet Express Passenger (dedicated CNG) CNG
## 11594 GMC Savana Cargo (dedicated CNG) CNG
## 11595 GMC Savana Passenger (dedicated CNG) CNG
## 12815 Dodge Caravan/Grand Caravan 2WD Gasoline or E85
## 12816 Chrysler Voyager/Town and Country 2WD Gasoline or E85
## 21506 Porsche 924 S Regular
## atvType
## 8128 Bifuel (CNG)
## 8129 Bifuel (CNG)
## 8130 Bifuel (LPG)
## 8131 Bifuel (LPG)
## 9175 CNG
## 9176 CNG
## 9184 Bifuel (CNG)
## 9185 Bifuel (CNG)
## 9186 Bifuel (LPG)
## 9187 Bifuel (LPG)
## 10283 Bifuel (LPG)
## 10284 Bifuel (LPG)
## 11585 Bifuel (LPG)
## 11586 Bifuel (LPG)
## 11587 Bifuel (CNG)
## 11588 Bifuel (CNG)
## 11589 Bifuel (CNG)
## 11590 Bifuel (CNG)
## 11592 CNG
## 11593 CNG
## 11594 CNG
## 11595 CNG
## 12815 FFV
## 12816 FFV
## 21506Independent Variables
Numeric
atvType
Definition
- type of alternative fuel or advanced technology vehicle
Remarks
- Missing values: The atvType information for majority 38248(91%) of the vehicles is Not Avilable. Replacing NA’s to Not Available for possible use in modeling.
## summary.vhcls_data.atvType.
## Bifuel (CNG) 20
## Bifuel (LPG) 8
## CNG 50
## Diesel 1070
## EV 168
## FFV 1412
## Hybrid 539
## Not Available 36707
## Plug-in Hybrid 107- Remaining 3708 vehicles plot on time scale exhibit three trends
- Sharp drop in diesel vehicles to the lowest in late nineties and slight pick-up later
- A strong rise in FFV (dual fuel vehicles) during 2000-2010 period and the a sharp drop to the lowest
## [1] Diesel CNG EV Bifuel (CNG) FFV
## [6] Hybrid Bifuel (LPG) Plug-in Hybrid
## 9 Levels: Bifuel (CNG) Bifuel (LPG) CNG Diesel EV FFV Hybrid ... Plug-in Hybrid
Post 2010, a consistent rising trend of Hybrid fuel and EV vehicles.
Looking at idividual etvTypes w.r.t. UCity, the Hybrid fuel and EV vehicles are strong predictors.
- Depending on the limited number of vehicles, it’s not safe to generalize above trends to all vehicels. Therefore, I’m inclined to drop this attribute as it as but with NA’s treatment, it has significant prediction power.
barrelsA08
Definitions
- barrels08 - annual petroleum consumption in barrels for fuelType1
- barrelsA08 - annual petroleum consumption in barrels for fuelType2
Remarks
- The median annual petroleum consumption (barrels08) is start to decrease post 2010.
chargeXXXX
- Definition
- charge120 : time to charge an electric vehicle in hours at 120 V
- charge240 : time to charge an electric vehicle in hours at 240 V
- charge240b : time to charge an electric vehicle in hours at 240 V using the alternate charger
- Remarks
- There are no vehicles with 120 V charging. Drop this variable.
- charge240: 252 vehicles. All non-EVs (39829) are having zero value for this variable.
- charge240: 62 vehicles. All non-EVs (39829) are having zero value for this variable.
- Due to nearly zero variance of charge columns, i am inclined to drop them off from model.
- Just out of interest; the correlation between UCity and charge240 ?
## Min. 1st Qu. Median Mean 3rd Qu. Max. ## 0 0 0 0 0 0
## [1] 252## [1] 62## Min. 1st Qu. Median Mean 3rd Qu. Max. ## 0.000000 0.000000 0.000000 0.007497 0.000000 8.500000
cityXXXX
- Definitions
- city08 : city MPG for fuelType1
- city08U : unrounded city MPG for fuelType1
- cityA08 : city MPG for fuelType2
- cityA08U : unrounded city MPG for fuelType2
- cityCD - city gasoline consumption (gallons/100 miles) in charge depleting mode (4)
- cityE - city electricity consumption in kw-hrs/100 miles
- cityUF - EPA city utility factor (share of electricity) for PHEV
- phevCity - EPA composite gasoline-electricity city MPGe for plug-in hybrid vehicles
- Remarks
- city08 is strongly correlated (0.9971) with UCity. Since UCity is actually unadjusted city MPG for fuelType1, including this variable in the model means to including UCity in the independent variables.
Such variables are fixation variables which produce too good to be true results. However the predictions are not actually predictions but a 1 to 1 mapping table. drop city08 from the model.
other city mpg variables are mostly present for post 2010 (for EV and Hybrid vehicles)
##
## Call:
## lm(formula = UCity ~ phevCity, data = vhcls_data)
##
## Coefficients:
## (Intercept) phevCity
## 22.9274 0.4425co2XXXX
- Definition
- co2 - tailpipe CO2 in grams/mile for fuelType1
- co2A - tailpipe CO2 in grams/mile for fuelType2
- co2TailpipeAGpm - tailpipe CO2 in grams/mile for fuelType2
- co2TailpipeGpm- tailpipe CO2 in grams/mile for fuelType1
- Remarks
- 31954 vehicles don’t have co2 information. Represented as -1.
- Few vehicles pre-2013 have co2 information
- Premium and Regular fuelType vehicle are top co2 emitters
## [1] 31954
- Interestingly, 168 EVs also reporting co2.
- But 168 EVs report zero co2, just boasting ?
- Correlation between UCity and co2 is strongly negative. (-0.75). Which means, at least for the 10k~ vehicles in the dataset, higher emission levels give lower mpg. More energy less miles.
## NULL
cylinders
- Definition
- engine cylinders
- Remarks
- 171 NA’s
## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 2.000 4.000 6.000 5.722 6.000 16.000 171- Treated as continuous variable which makes no sense. Type should be cast to categorical.
- 2 3 4 5 6 8 10 12 16 cylender vehicles.
- most vehicles (30k+) have 4, 6, 8 cylinders
## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 2.000 4.000 6.000 5.722 6.000 16.000 171
## UCity cylinders
## UCity 1.0000 -0.8157
## cylinders -0.8157 1.0000- cylinders are negatively correlated with UCity. (-0.8157), which indicates the bigger the vehicle the less the miles per gallon. Make sense.
- 4, 6 and 8 cylinders vehicles are getting slightly better on UCity in recent years
displ
Definition
- Displacement, combined volume of the pistons inside the cylinders
Remarks
- negatively correlated to UCity (spearman, -0.846), which makes sense. The larger the engine (more cylinders), the less the UCity/MPG.
displ variable has missing values for the EV vehicles. It should be like this.
Replacing missing values with ‘NULL’ and making dspl as a factor is going to help make better sense of this variable.
One Odd Case is of an EV vehicle with displacement value zero. Making it ‘NULL’ is necessary.
Alternatively replace all NA’s with zero. That is consisstent with reality. EV has zero displacement. In addition, DT/LM will be happier with displ being numeric variable.
over the years, displ distributions is not changing noticeably.
evMotor
- Definition
- electric motor (kw-hrs)
- Remarks
- 141 types of motors,
- 38345 vehicles don’t have any
- 736 EV, Hybrid, Plug-in Hybrid atvType vehicles have evMotors
- A strong overfitting predictor for high UCity
- evMotors like ‘88 kW AC PMSM’ are outperforming UCity measure, including such variable into model can lead to overfitting.
## [1] EV Hybrid Plug-in Hybrid
## 9 Levels: Bifuel (CNG) Bifuel (LPG) CNG Diesel EV FFV Hybrid ... Plug-in Hybrid## [1] 736##
## Call:
## lm(formula = UCity ~ evMotor_Bin, data = vhcls_data)
##
## Coefficients:
## (Intercept) evMotor_BinnoMotor
## 64.48 -42.27##
## Call:
## lm(formula = UCity ~ evMotor, data = vhcls_data)
##
## Coefficients:
## (Intercept)
## 22.2056
## evMotor100 kW AC PMSM
## 154.7944
## evMotor100 kW DCPM
## 88.0944
## evMotor101V Ni-MH
## 27.1932
## evMotor102kW AC Induction
## 57.3638
## evMotor104 kW AC Induction
## 160.9944
## evMotor104 kW ACPM
## 160.9944
## evMotor105 kW ACPM
## 160.9944
## evMotor107 kW AC Induction
## 134.9944
## evMotor107 kW AC PMSM
## 138.9982
## evMotor110 kW DCPM
## 156.3944
## evMotor111 kW
## 24.8224
## evMotor111 kW 3-Phase AC
## 25.3944
## evMotor115 kW AC Induction
## 89.8348
## evMotor115V Li-Ion
## 9.3947
## evMotor120 kW AC PMSM
## 157.3944
## evMotor124 kW
## 42.6845
## evMotor125 kW AC Induction
## 136.2160
## evMotor125kW
## 30.1611
## evMotor125kW AC Induction
## 26.9944
## evMotor126 kW
## 25.3944
## evMotor126 kW 3-Phase AC
## 25.3944
## evMotor126V Li-Ion
## 2.5620
## evMotor132 kW AC Induction
## 99.4944
## evMotor132 kW AC Synchronous
## -2.8466
## evMotor132kW
## -2.8466
## evMotor135kW AC PMSM
## 39.9944
## evMotor140 (front) 140 (rear) (70 kW-hr battery pack)
## 111.4944
## evMotor140 (front) 140 (rear) (85 kW-hr battery pack)
## 104.1944
## evMotor140 (front) 140 (rear) (90 kW-hr battery pack)
## 104.1944
## evMotor144V Li-Ion
## 20.1248
## evMotor144V Ni-MH
## 48.7900
## evMotor147 and 188 kW AC 3-Phase
## 149.0944
## evMotor147 and 211 kW AC 3-Phase
## 149.0944
## evMotor150 and 150 kw DCPM
## 2.2944
## evMotor150 kW
## 124.6443
## evMotor150 kW ACPM
## 159.9944
## evMotor158V Ni-MH
## 32.3944
## evMotor16 and 37 kW AC Induction
## 56.6141
## evMotor164 (front) 350 (rear) (85 kW-hr battery pack)
## 98.8944
## evMotor164 (front) 350 (rear) (90 kW-hr battery pack)
## 98.8944
## evMotor18 kW
## 48.6944
## evMotor18 kW AC Induction
## 48.6944
## evMotor180 and 350 kW AC Induction (85 kW-hr battery pack)
## 95.5944
## evMotor192 kW AC 3-Phase
## 164.4944
## evMotor193 (front) 193 (rear) (100 kW-hr battery pack)
## 103.3944
## evMotor193 (front) 193 (rear) (60 kW-hr battery pack)
## 108.6944
## evMotor193 (front) 193 (rear) (70 kW-hr battery pack)
## 111.4944
## evMotor193 (front) 193 (rear) (75 kW-hr battery pack)
## 108.8444
## evMotor193 (front) 193 (rear) (85 kW-hr battery pack)
## 104.1944
## evMotor193 (front) 193 (rear) (90 kW-hr battery pack)
## 106.3694
## evMotor193 (front) 375 (rear) (100 kW-hr battery pack)
## 96.1944
## evMotor193 (front) 375 (rear) (85 kW-hr battery pack)
## 98.8944
## evMotor193 (front) 375 (rear) (90 kW-hr battery pack)
## 100.1194
## evMotor2 @ 150 kw (300 kw)
## 2.6904
## evMotor202V Li-Ion
## 21.1809
## evMotor202V Ni-MH
## 43.9937
## evMotor207V Li-Ion
## 59.3625
## evMotor211 kW AC 3-Phase
## 172.7944
## evMotor220V Li-Ion
## 35.2944
## evMotor222 kW AC PMSM
## 54.3729
## evMotor225 kW AC Induction
## 95.7444
## evMotor225 kW AC Induction (60 kW-hr battery pack)
## 96.3944
## evMotor24 KW AC Synchronous
## 98.1515
## evMotor240V Li-Ion
## 46.1989
## evMotor245V Ni-MH
## 29.7115
## evMotor259V Li-Ion
## 32.9202
## evMotor260 kW AC Induction
## 88.1944
## evMotor260V Li-Ion
## 11.7453
## evMotor266V Li-Ion
## 8.1944
## evMotor27 KW AC Induction
## 83.1069
## evMotor270 kW AC Induction
## 88.1944
## evMotor270 kW AC Induction (75 kW-hr battery pack)
## 109.8944
## evMotor270 kW AC Induction (85 kW-hr battery pack)
## 88.1944
## evMotor270V Li-Ion
## 27.4024
## evMotor275V Ni-MH
## 34.7921
## evMotor280V Li-Ion
## 37.6450
## evMotor285 kW AC Induction (60 kW-hr battery pack)
## 109.9944
## evMotor285 kW AC Induction (70 kW-hr battery pack)
## 88.1944
## evMotor285 kW AC Induction (75 kW-hr battery pack)
## 109.8944
## evMotor285 kW AC Induction (85 kW-hr battery pack)
## 88.1944
## evMotor285 kW AC Induction (90 kW-hr battery pack)
## 88.1944
## evMotor288V Ni-MH
## 8.2222
## evMotor30 kW DCPM
## 112.4214
## evMotor300V Li-Ion
## 43.6944
## evMotor300V Ni-MH
## 2.5944
## evMotor311V Li-Ion
## 11.5062
## evMotor312V Ni-MH
## -1.4433
## evMotor32kW IPMSM
## 45.5444
## evMotor330V Ni-MH
## 19.8865
## evMotor34 and 65kW 3-phase Sync.
## 10.7444
## evMotor34 and 65kW DCPM
## 8.1176
## evMotor346V Li-Ion
## 13.8309
## evMotor36V Ni-MH
## 9.8194
## evMotor374V Li-Ion
## 7.2686
## evMotor48 and 87 kW 3-Phase AC
## 36.4944
## evMotor48V Li-Ion
## 2.5415
## evMotor49 kW DC Brushless
## 46.5699
## evMotor49 kW DCPM
## 156.7170
## evMotor49kW DC Brushless
## 46.5699
## evMotor50 KW DC
## 98.3052
## evMotor50 kW IPMSM
## 29.0277
## evMotor50kW IPMSM
## 31.9944
## evMotor52 kW AC Induction
## 66.1944
## evMotor55 and 174kW 3-Phase AC
## 20.5944
## evMotor55 kW DCPM
## 151.5944
## evMotor56kW AC Induction
## 28.0929
## evMotor59 and 61 kW 3-Phase AC
## 9.4944
## evMotor59 and 61kW 3-Phase AC
## 9.4944
## evMotor60 KW AC Synchronous
## 151.6444
## evMotor60 kW DCPM
## 15.4034
## evMotor62 KW AC Induction
## 94.0013
## evMotor65 kW AC Induction
## 13.4944
## evMotor65kW
## 14.2802
## evMotor67 KW AC
## 59.3398
## evMotor67 KW AC Induction
## 66.4786
## evMotor67 KW AC Induction
## 40.2018
## evMotor68 kW
## 32.7944
## evMotor68 kW DCPM
## 34.3801
## evMotor70 kW
## 6.6944
## evMotor70 kW DC Brushless
## 3.4527
## evMotor75 kW AC PM/B
## 82.2944
## evMotor75 kW AC PMSM
## 64.8509
## evMotor80 kW DCPM
## 108.3424
## evMotor81 kW AC PMSM
## 150.1194
## evMotor82 kW AC Induction
## 151.1833
## evMotor82 kW ACIPM
## 152.6499
## evMotor83 kW AC Induction
## 7.1944
## evMotor83kW DCPM
## 12.3598
## evMotor85 kW AC Induction
## 33.1644
## evMotor85 kW AC PMSM
## 157.1444
## evMotor85 kW DCPM
## 6.8230
## evMotor86V Li-Ion
## 0.6128
## evMotor88 kW AC PMSM
## 202.5944
## evMotor89 kW AC Induction
## 25.0733
## evMotor92 kW DCPM
## 166.2031
## evMotor95 kW and 116 kW DC Brushless
## 6.3944
## evMotor96kW
## 16.1944
## evMotor96kW AC Induction
## 12.6611
## evMotorTwo 444V Li-Ion
## -8.1134
## evMotorTwo 480V Li-Ion
## -7.1886feScore
- Definition
- feScore: EPA Fuel Economy Score,
- Remarks
- available only for 8054 post-2012 vehicles.
- 32027 Pre-(including) 2012 vihicles feScore is -1
- Although numeric, better to treat it as categorical variable
## -1 1 2 3 4 5 6 7 8 9 10
## 32027 164 371 851 1541 2095 1195 938 552 118 229## [1] 8054
## [1] 0.3978345
combXXXX
- Definitions
- comb08 : combined MPG for fuelType1
- comb08U : unrounded combined MPG for fuelType1
- combA08 : combined MPG for fuelType2
- combA08U : unrounded combined MPG for fuelType2
- combE : combined electricity consumption in kw-hrs/100 miles
- combinedCD : combined gasoline consumption (gallons/100 miles) in charge depleting mode (4)
- combinedUF : EPA combined utility factor (share of electricity) for PHEV
- Remarks
- comb08 strongly correlated (0.9839) with UCity
- Other comb variables are limited to last few year.
fuelCostXX
Definition
- fuelCost08 - annual fuel cost for fuelType1
- fuelCostA08 - annual fuel cost for fuelType2
Remarks
- As expected, fuelCost is strongly correlated with barrels08 (0.9256)
- Post2012 the median fuelCost08 is starting to decrease slowly but noticeably
- fuelCostA08 pre 2000 is non-existent.
- EV, Hybrid and Plug-in Hybrid are least expensive i.t.o fuelCost08
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 500 1950 2350 2378 2700 7350fuelTypeX
Definition
- fuelType : fuel type with fuelType1 and fuelType2 (if applicable)
- fuelType1 : fuel type 1. For single fuel vehicles, this will be the only fuel. For dual fuel vehicles, this will be the conventional fuel.
- fuelType2 : fuel type 2. For dual fuel vehicles, this will be the alternative fuel (e.g. E85, Electricity, CNG, LPG). For single fuel vehicles, this field is not used
## Warning: The shape palette can deal with a maximum of 6 discrete values because
## more than 6 becomes difficult to discriminate; you have 14. Consider
## specifying shapes manually if you must have them.## Warning: Removed 118 rows containing missing values (geom_point).
ghgScoreX
Definition
- EPA GHG score for dual fuel vehicle running on the alternative fuel (-1 = Not available)
Remarks
- just like feScore, there are 32027 vehicles without ghgScore (represented as -1)
- Available for onle 8054 vehicles
- Both feScore and ghgScore are almost copies of each other (cor 0.99),
- 32027 vehicles without ghgScore
- 9990 vehicles with ghgScore available
## -1 1 2 3 4 5 6 7 8 9 10
## 32027 163 371 855 1570 2081 1196 962 491 116 249## [1] 8054- Over the year ghgScore progression
- ghgScore ~ UCity
## [1] 0.3973776
- drop candidate.
highwayXXXX
Definitions
- highway08 - highway MPG for fuelType1
- highway08U - unrounded highway MPG for fuelType1
- highwayA08 - highway MPG for fuelType2
- highwayA08U - unrounded highway MPG for fuelType2
- highwayCD - highway gasoline consumption (gallons/100miles) in charge depleting mode
- highwayE - highway electricity consumption in kw-hrs/100 miles
- highwayUF - EPA highway utility factor (share of electricity) for PHEV
- UHighway - unadjusted highway MPG for fuelType1; see the description of the EPA test procedures
- UHighwayA - unadjusted highway MPG for fuelType2; see the description of the EPA test procedures
- Remarks
- highway08, UHighway strongly correlated (0.9257, 0.9245) with UCity
- highway08, UHighway are 0.99 correlated
- other highway variables are limited to last f
volume hlv, hpv, lv2, lv4, pv2, pv4
- Definition
- hlv - hatchback luggage volume (cubic feet)
- hpv - hatchback passenger volume (cubic feet)
- lv2 - 2 door luggage volume (cubic feet)
- lv4 - 4 door luggage volume (cubic feet)
- pv2 - 2-door passenger volume (cubic feet)
- pv4 - 4-door passenger volume (cubic feet)
Remarks
- 35240 vehicles don’t have hlv, hpc, lv2, lv4 information (zero)
- 35531 vehicles don’t have pv2, pv4 information (zero)
- hlv~hpv 0.933
- There is a mild correlation between UCity and hlv or hpv
- There is very weak correlation between UCity and lv2, lv4, pv2, pv4
- Drop cadidate
range
Definition - range NOT in the dictionary but present in the data
Remarks
- 39913 vehicles don’t have range information (rep as zero)
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.6164 0.0000 335.0000- For remaining 168 vehicles range is strongly correlated with UCity (0.60)
- The correlation is not reliable because the distribution is heteroskedastic
## [1] 168## [1] 0.6013586
- The 168 vehicles with range value are all EV
## [1] EV
## 9 Levels: Bifuel (CNG) Bifuel (LPG) CNG Diesel EV FFV Hybrid ... Plug-in HybridrangeXXX
Definition
- rangeA - EPA range for fuelType2
- rangeCityA - EPA city range for fuelType2
- rangeHwyA - EPA highway range for fuelType2
Remarks
- 38539 vehicles don’t have rangeA values ""
- Rest 1542 vehicles are atvType in (Bifuel (CNG), FFV, Bifuel (LPG) Plug-in Hybrid))
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.6164 0.0000 335.0000## [1] 0## range UCity
## range 1.0000000 0.6013586
## UCity 0.6013586 1.0000000
youSaveSpend
Definition
- you save/spend over 5 years compared to an average car ($)
Remarks
- Storngly correlated with UCity (0.6583)
- youSaveSpend also shows a trend reversal at year 2010
- EV, Hybrid, Plug-in Hybrid and CNG vechicles return positive
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -29000 -5750 -4000 -4135 -2000 5250## youSaveSpend UCity
## youSaveSpend 1.000000 0.658371
## UCity 0.658371 1.000000
Categorical + Logical
cylinders_cat
- Definition
- factorized cylinders variable
drive
Definition
- drive - drive axle type
Remarks
- 1189 vehicle have missing values for drive, replaced with NA
- Different axle type have mixed relation with UCity. No clear indication of mutual information. cor(-0.117)
## 2-Wheel Drive
## 1189 507
## 4-Wheel Drive 4-Wheel or All-Wheel Drive
## 1328 6648
## All-Wheel Drive Automatic (A1)
## 2713 1
## Front-Wheel Drive Part-time 4-Wheel Drive
## 13939 217
## Rear-Wheel Drive
## 13539
## [1] 40081## [1] "UCity~drive correlation -0.1171"
feScore_cat
- Definition
- factorized cylinders variable
guzzler
- Definition
- guzzler- if G or T, this vehicle is subject to the gas guzzler tax
- Remarks
- S value stand for guzzler tax exempt ?
- 37704 vehicles have missing values for guzzler (does that mean they are exempt from guzzler tax?)
- Remaining 2377 vehicles are FFV and Hybrid atvType vehicles
- As expected, guzzler tax paying vehicles have lower-end UCity
## G S T
## 37704 1398 15 964
## [1] Not Available FFV Hybrid
## 9 Levels: Bifuel (CNG) Bifuel (LPG) CNG Diesel EV FFV Hybrid ... Plug-in Hybrid
make, mfrCode
- Definition
- make - manufacturer (division)
- mfrCode - 3-character manufacturer code
- Remarks
- both variables contain same information. Either of them can be dropped.
## [1] Alfa Romeo Ferrari
## [3] Dodge Ford
## [5] GMC Toyota
## [7] Volkswagen AM General
## [9] Chevrolet E. P. Dutton, Inc.
## [11] Grumman Olson Jeep
## [13] Plymouth American Motors Corporation
## [15] Isuzu Mitsubishi
## [17] Subaru Suzuki
## [19] Nissan Honda
## [21] Mercedes-Benz Mazda
## [23] Maserati Aston Martin
## [25] Audi Chrysler
## [27] Mercury Pontiac
## [29] Buick Cadillac
## [31] Oldsmobile Volvo
## [33] Jaguar Bertone
## [35] Lotus Renault
## [37] Rolls-Royce BMW
## [39] Pininfarina Bill Dovell Motor Car Company
## [41] Porsche TVR Engineering Ltd
## [43] Merkur Peugeot
## [45] Saab Lincoln
## 135 Levels: Acura Alfa Romeo AM General ... Yugo
##
## Call:
## lm(formula = UCity ~ mfrCode, data = vhcls_data)
##
## Coefficients:
## (Intercept) mfrCodeADX mfrCodeASX mfrCodeAZD mfrCodeBEX mfrCodeBGT
## 21.5954 1.4108 -5.0428 66.8046 -7.0454 -11.5954
## mfrCodeBMX mfrCodeBYD mfrCodeCDA mfrCodeCRX mfrCodeDSX mfrCodeFEX
## 6.5156 74.1828 88.7046 3.8127 2.3934 -5.6543
## mfrCodeFJX mfrCodeFMX mfrCodeFSK mfrCodeFTG mfrCodeGMX mfrCodeHNX
## 7.3045 4.6702 3.3006 7.1046 2.1571 12.2258
## mfrCodeHYX mfrCodeJCX mfrCodeJLX mfrCodeKAL mfrCodeKGG mfrCodeKMX
## 9.7952 -2.0571 1.3929 2.9046 -8.2954 10.5924
## mfrCodeLRX mfrCodeLTX mfrCodeMAX mfrCodeMBV mfrCodeMBX mfrCodeMLN
## -4.4330 0.6201 -3.7190 -5.4454 5.2703 -2.0218
## mfrCodeMTX mfrCodeNLX mfrCodeNSX mfrCodePGN mfrCodePRX mfrCodeQTM
## 15.2185 -7.9738 6.3905 -8.6954 1.9600 -0.8954
## mfrCodeRII mfrCodeRRG mfrCodeSAX mfrCodeSKR mfrCodeSKX mfrCodeTKX
## -5.7110 -7.1661 2.3880 -4.4944 6.2741 10.8277
## mfrCodeTSL mfrCodeTVP mfrCodeTYX mfrCodeVGA mfrCodeVVX mfrCodeVWX
## 105.9098 -6.2187 9.2301 5.7141 4.9026 7.5194mpgData
- Definition
- has My MPG data
- Remarks
- 12714 vehicles have mpgData Y
- Not a strong predictor for UCity
## [1] 12714
phevBlended
Definition - phevBlended - if true, this vehicle operates on a blend of gasoline and electricity in charge depleting mode Remarks - 40005 N, 76 Y - For 76 Y vehicles, the phevBlended is a strong predictor
## false true
## 40005 76##
## Call:
## lm(formula = UCity ~ phevBlended, data = vhcls_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -23.867 -4.843 -1.710 2.678 201.857
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 22.94334 0.05218 439.71 <2e-16 ***
## phevBlendedtrue 20.28253 1.19827 16.93 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 10.44 on 40079 degrees of freedom
## Multiple R-squared: 0.007098, Adjusted R-squared: 0.007073
## F-statistic: 286.5 on 1 and 40079 DF, p-value: < 2.2e-16trans_dscr
Definition
- trans_dscr - transmission descriptor;
Remarks
- 25034 vehicles have missing values for this variable
- SIL EMS transmission type is most important one (lm coef 33)
- These are Honda Civic and Insight cars (2004, 2005, 2006)
## 2LKUP 2MODE 2MODE 2LKUP 2MODE 3LKUP
## 25034 63 448 383 22
## 2MODE CLKUP 2MODE CLKUP FW 2MODE DC/FW 2MODE VLKUP 3LKUP
## 1235 2 19 9 15
## 3MODE 3MODE 2LKUP 3MODE CLKUP 3MODE CLKUP FW 3MODE VLKUP
## 166 2 517 1 5
## 4MODE 4MODE CLKUP 6MODE 6MODE CLKUP CLKUP
## 35 6 6 41 7809
## CMODE CMODE CLKUP CMODE VLKUP Creeper DC/FW
## 150 130 17 525 53
## Elec Overdrive EMS EMS 2MODE EMS 2MODE CLKU EMS 2MODE CLKUP
## 3 252 46 5 520
## EMS 3MODE EMS 3MODE CLKUP EMS 5MODE EMS CLKUP EMS CMODE CLKUP
## 11 3 1 50 2
## fuel injection Lockup Lockup A3 LONG RATIO Mech Overdrive
## 1 9 1 2 3
## Overdrive SIL SIL 2MODE CLKUP SIL 3MODE SIL 3MODE CLKUP
## 10 2189 5 6 2
## SIL CLKUP SIL CMODE SIL Creeper SIL EMS VLKUP
## 1 2 72 7 52
## VMODE VMODE CLKUP VMODE VLKUP
## 2 105 26##
## Call:
## lm(formula = UCity ~ trans_dscr, data = vhcls_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -24.427 -4.427 -1.123 2.573 200.373
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 24.42665 0.06457 378.320 < 2e-16 ***
## trans_dscr2LKUP -2.16581 1.28868 -1.681 0.092841 .
## trans_dscr2MODE -4.98120 0.48695 -10.229 < 2e-16 ***
## trans_dscr2MODE 2LKUP -3.30358 0.52598 -6.281 3.40e-10 ***
## trans_dscr2MODE 3LKUP -1.49737 2.17896 -0.687 0.491964
## trans_dscr2MODE CLKUP -3.28028 0.29778 -11.016 < 2e-16 ***
## trans_dscr2MODE CLKUP FW -3.42665 7.22390 -0.474 0.635253
## trans_dscr2MODE DC/FW -8.90347 2.34454 -3.798 0.000146 ***
## trans_dscr2MODE VLKUP -6.48838 3.40585 -1.905 0.056779 .
## trans_dscr3LKUP -0.93035 2.63848 -0.353 0.724383
## trans_dscr3MODE -4.60658 0.79552 -5.791 7.06e-09 ***
## trans_dscr3MODE 2LKUP -8.59330 7.22390 -1.190 0.234224
## trans_dscr3MODE CLKUP -3.69621 0.45390 -8.143 3.96e-16 ***
## trans_dscr3MODE CLKUP FW -8.77805 10.21593 -0.859 0.390207
## trans_dscr3MODE VLKUP -0.10665 4.56907 -0.023 0.981377
## trans_dscr4MODE -4.61372 1.72798 -2.670 0.007588 **
## trans_dscr4MODE CLKUP -6.77939 4.17105 -1.625 0.104098
## trans_dscr6MODE -6.40625 4.17105 -1.536 0.124575
## trans_dscr6MODE CLKUP -9.42931 1.59673 -5.905 3.55e-09 ***
## trans_dscrCLKUP -4.58441 0.13241 -34.622 < 2e-16 ***
## trans_dscrCMODE -4.67248 0.83661 -5.585 2.35e-08 ***
## trans_dscrCMODE CLKUP -2.60704 0.89830 -2.902 0.003708 **
## trans_dscrCMODE VLKUP 1.89361 2.47852 0.764 0.444866
## trans_dscrCreeper -8.43588 0.45050 -18.726 < 2e-16 ***
## trans_dscrDC/FW -1.88634 1.40472 -1.343 0.179325
## trans_dscrElec Overdrive -3.42665 5.89841 -0.581 0.561280
## trans_dscrEMS -0.50839 0.64676 -0.786 0.431844
## trans_dscrEMS 2MODE -6.88840 1.50761 -4.569 4.91e-06 ***
## trans_dscrEMS 2MODE CLKU -7.29999 4.56907 -1.598 0.110118
## trans_dscrEMS 2MODE CLKUP -6.40862 0.45262 -14.159 < 2e-16 ***
## trans_dscrEMS 3MODE -12.33574 3.08083 -4.004 6.24e-05 ***
## trans_dscrEMS 3MODE CLKUP -11.18679 5.89841 -1.897 0.057891 .
## trans_dscrEMS 5MODE 2.67335 10.21593 0.262 0.793566
## trans_dscrEMS CLKUP -9.17839 1.44616 -6.347 2.22e-10 ***
## trans_dscrEMS CMODE CLKUP -4.73075 7.22390 -0.655 0.512552
## trans_dscrfuel injection -5.53775 10.21593 -0.542 0.587773
## trans_dscrLockup -7.05285 3.40585 -2.071 0.038384 *
## trans_dscrLockup A3 -9.42665 10.21593 -0.923 0.356148
## trans_dscrLONG RATIO -11.47665 7.22390 -1.589 0.112135
## trans_dscrMech Overdrive -4.75999 5.89841 -0.807 0.419674
## trans_dscrOverdrive -7.95998 3.23114 -2.464 0.013762 *
## trans_dscrSIL -0.23864 0.22769 -1.048 0.294615
## trans_dscrSIL 2MODE CLKUP 13.89321 4.56907 3.041 0.002362 **
## trans_dscrSIL 3MODE -10.68207 4.17105 -2.561 0.010441 *
## trans_dscrSIL 3MODE CLKUP -10.02735 7.22390 -1.388 0.165120
## trans_dscrSIL CLKUP -0.41005 10.21593 -0.040 0.967983
## trans_dscrSIL CMODE -8.42665 7.22390 -1.166 0.243421
## trans_dscrSIL Creeper -8.90660 1.20566 -7.387 1.53e-13 ***
## trans_dscrSIL EMS 33.31620 3.86172 8.627 < 2e-16 ***
## trans_dscrVLKUP 5.69771 1.41814 4.018 5.89e-05 ***
## trans_dscrVMODE -6.38575 7.22390 -0.884 0.376714
## trans_dscrVMODE CLKUP -1.09962 0.99904 -1.101 0.271046
## trans_dscrVMODE VLKUP 6.48897 2.00451 3.237 0.001208 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 10.22 on 40028 degrees of freedom
## Multiple R-squared: 0.04984, Adjusted R-squared: 0.04861
## F-statistic: 40.38 on 52 and 40028 DF, p-value: < 2.2e-16trany
Definition
- transmission type
Remarks
- Most common transmission type is Automatic 4 and 4 Speed
- Automatic (A1) is the only positive coef (48) with global mean 88
- All Automatic (A1) vehicles are EV (including best UCity Hyundai vehicle Ioniq) except one Plug-in Hybrid BMW
- Over the years, Automatic transmission cars are beoming more and more common
- On broader categorization (Automatic, Manual), Manual transmission does improve UCity 1.43922 as much as.
## Automatic (A1)
## 11 163
## Automatic (AM-S6) Automatic (AM-S7)
## 113 374
## Automatic (AM-S8) Automatic (AM-S9)
## 40 2
## Automatic (AM5) Automatic (AM6)
## 14 136
## Automatic (AM7) Automatic (AM8)
## 216 6
## Automatic (AV-S10) Automatic (AV-S6)
## 5 189
## Automatic (AV-S7) Automatic (AV-S8)
## 126 43
## Automatic (L3) Automatic (L4)
## 2 2
## Automatic (S10) Automatic (S4)
## 69 233
## Automatic (S5) Automatic (S6)
## 830 2984
## Automatic (S7) Automatic (S8)
## 307 1421
## Automatic (S9) Automatic (variable gear ratios)
## 65 766
## Automatic 10-spd Automatic 3-spd
## 16 3151
## Automatic 4-spd Automatic 5-spd
## 11045 2198
## Automatic 6-spd Automatic 7-spd
## 1579 708
## Automatic 8-spd Automatic 9-spd
## 347 208
## Manual 3-spd Manual 4-spd
## 77 1483
## Manual 4-spd Doubled Manual 5-spd
## 17 8351
## Manual 6-spd Manual 7-spd
## 2671 113
##
## Call:
## lm(formula = UCity ~ trany, data = vhcls_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -112.656 -3.662 -0.579 2.886 159.877
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 88.885 2.017 44.07 <2e-16 ***
## tranyAutomatic (A1) 48.271 2.084 23.16 <2e-16 ***
## tranyAutomatic (AM-S6) -58.007 2.113 -27.46 <2e-16 ***
## tranyAutomatic (AM-S7) -65.990 2.046 -32.25 <2e-16 ***
## tranyAutomatic (AM-S8) -62.173 2.277 -27.30 <2e-16 ***
## tranyAutomatic (AM-S9) -60.185 5.142 -11.71 <2e-16 ***
## tranyAutomatic (AM5) -44.123 2.695 -16.37 <2e-16 ***
## tranyAutomatic (AM6) -53.657 2.097 -25.59 <2e-16 ***
## tranyAutomatic (AM7) -67.183 2.068 -32.49 <2e-16 ***
## tranyAutomatic (AM8) -60.602 3.395 -17.85 <2e-16 ***
## tranyAutomatic (AV-S10) -52.199 3.608 -14.47 <2e-16 ***
## tranyAutomatic (AV-S6) -54.206 2.075 -26.13 <2e-16 ***
## tranyAutomatic (AV-S7) -53.421 2.103 -25.40 <2e-16 ***
## tranyAutomatic (AV-S8) -58.879 2.260 -26.05 <2e-16 ***
## tranyAutomatic (L3) -71.212 5.142 -13.85 <2e-16 ***
## tranyAutomatic (L4) -71.945 5.142 -13.99 <2e-16 ***
## tranyAutomatic (S10) -66.174 2.172 -30.47 <2e-16 ***
## tranyAutomatic (S4) -66.279 2.064 -32.11 <2e-16 ***
## tranyAutomatic (S5) -67.540 2.030 -33.27 <2e-16 ***
## tranyAutomatic (S6) -65.364 2.021 -32.35 <2e-16 ***
## tranyAutomatic (S7) -67.521 2.053 -32.89 <2e-16 ***
## tranyAutomatic (S8) -65.048 2.025 -32.13 <2e-16 ***
## tranyAutomatic (S9) -63.524 2.181 -29.13 <2e-16 ***
## tranyAutomatic (variable gear ratios) -51.185 2.031 -25.20 <2e-16 ***
## tranyAutomatic 10-spd -71.121 2.620 -27.15 <2e-16 ***
## tranyAutomatic 3-spd -66.926 2.020 -33.12 <2e-16 ***
## tranyAutomatic 4-spd -69.027 2.018 -34.21 <2e-16 ***
## tranyAutomatic 5-spd -69.671 2.022 -34.46 <2e-16 ***
## tranyAutomatic 6-spd -67.688 2.024 -33.45 <2e-16 ***
## tranyAutomatic 7-spd -67.454 2.032 -33.19 <2e-16 ***
## tranyAutomatic 8-spd -68.289 2.049 -33.34 <2e-16 ***
## tranyAutomatic 9-spd -63.249 2.070 -30.56 <2e-16 ***
## tranyManual 3-spd -71.337 2.156 -33.09 <2e-16 ***
## tranyManual 4-spd -66.996 2.024 -33.09 <2e-16 ***
## tranyManual 4-spd Doubled -57.885 2.588 -22.36 <2e-16 ***
## tranyManual 5-spd -64.423 2.018 -31.92 <2e-16 ***
## tranyManual 6-spd -65.401 2.021 -32.36 <2e-16 ***
## tranyManual 7-spd -66.132 2.113 -31.30 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.689 on 40043 degrees of freedom
## Multiple R-squared: 0.5925, Adjusted R-squared: 0.5921
## F-statistic: 1573 on 37 and 40043 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = UCity ~ trany_new, data = vhcls_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -22.525 -4.725 -1.525 2.575 202.275
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 22.52494 0.06319 356.45 <2e-16 ***
## trany_newManual 1.43922 0.11216 12.83 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 10.45 on 40079 degrees of freedom
## Multiple R-squared: 0.004092, Adjusted R-squared: 0.004067
## F-statistic: 164.7 on 1 and 40079 DF, p-value: < 2.2e-16VClass
Definition
- EPA vehicle size class
Remarks
- Compact, Subcompact and Midsize cars are the top 3 in the data
- Special Purpose Verticia, Vans Passenger are the least
- Linear model does show the negative effect of vehicle class
- larger vehicles (Truck and Vans) the lower the UCity with global mean 26.5827
##
## Call:
## lm(formula = UCity ~ VClass, data = vhcls_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -24.209 -3.905 -1.054 1.939 199.746
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 26.5827 0.1300 204.541 < 2e-16
## VClassLarge Cars -3.8831 0.2525 -15.377 < 2e-16
## VClassMidsize Cars -1.5285 0.1938 -7.886 3.20e-15
## VClassMidsize Station Wagons -3.1773 0.4451 -7.138 9.60e-13
## VClassMidsize-Large Station Wagons -4.7331 0.4062 -11.653 < 2e-16
## VClassMinicompact Cars -2.3737 0.2966 -8.004 1.24e-15
## VClassMinivan - 2WD -5.5012 0.5383 -10.220 < 2e-16
## VClassMinivan - 4WD -7.2576 1.4285 -5.081 3.78e-07
## VClassSmall Pickup Trucks -3.4385 0.4443 -7.738 1.03e-14
## VClassSmall Pickup Trucks 2WD -3.7426 0.4790 -7.813 5.71e-15
## VClassSmall Pickup Trucks 4WD -5.9404 0.6533 -9.093 < 2e-16
## VClassSmall Sport Utility Vehicle 2WD 1.8148 0.4474 4.056 5.00e-05
## VClassSmall Sport Utility Vehicle 4WD -0.6477 0.3863 -1.677 0.0936
## VClassSmall Station Wagons 1.6988 0.2800 6.067 1.32e-09
## VClassSpecial Purpose Vehicle 0.2173 9.8566 0.022 0.9824
## VClassSpecial Purpose Vehicle 2WD -6.1868 0.4098 -15.096 < 2e-16
## VClassSpecial Purpose Vehicle 4WD -7.7128 0.5747 -13.421 < 2e-16
## VClassSpecial Purpose Vehicles -7.5729 0.2892 -26.184 < 2e-16
## VClassSpecial Purpose Vehicles/2wd -3.6395 6.9703 -0.522 0.6016
## VClassSpecial Purpose Vehicles/4wd -6.8959 6.9703 -0.989 0.3225
## VClassSport Utility Vehicle - 2WD -5.1484 0.2768 -18.603 < 2e-16
## VClassSport Utility Vehicle - 4WD -7.2713 0.2517 -28.885 < 2e-16
## VClassStandard Pickup Trucks -9.5215 0.2412 -39.483 < 2e-16
## VClassStandard Pickup Trucks 2WD -7.9801 0.3106 -25.696 < 2e-16
## VClassStandard Pickup Trucks 4WD -9.4305 0.3316 -28.443 < 2e-16
## VClassStandard Pickup Trucks/2wd -11.9551 4.9296 -2.425 0.0153
## VClassStandard Sport Utility Vehicle 2WD -6.1120 0.6533 -9.356 < 2e-16
## VClassStandard Sport Utility Vehicle 4WD -3.7400 0.4304 -8.690 < 2e-16
## VClassSubcompact Cars -0.5774 0.1902 -3.036 0.0024
## VClassTwo Seaters -3.9304 0.2547 -15.433 < 2e-16
## VClassVans -9.6943 0.3194 -30.351 < 2e-16
## VClassVans Passenger -10.4344 6.9703 -1.497 0.1344
## VClassVans, Cargo Type -10.8136 0.4885 -22.135 < 2e-16
## VClassVans, Passenger Type -11.5313 0.5712 -20.189 < 2e-16
##
## (Intercept) ***
## VClassLarge Cars ***
## VClassMidsize Cars ***
## VClassMidsize Station Wagons ***
## VClassMidsize-Large Station Wagons ***
## VClassMinicompact Cars ***
## VClassMinivan - 2WD ***
## VClassMinivan - 4WD ***
## VClassSmall Pickup Trucks ***
## VClassSmall Pickup Trucks 2WD ***
## VClassSmall Pickup Trucks 4WD ***
## VClassSmall Sport Utility Vehicle 2WD ***
## VClassSmall Sport Utility Vehicle 4WD .
## VClassSmall Station Wagons ***
## VClassSpecial Purpose Vehicle
## VClassSpecial Purpose Vehicle 2WD ***
## VClassSpecial Purpose Vehicle 4WD ***
## VClassSpecial Purpose Vehicles ***
## VClassSpecial Purpose Vehicles/2wd
## VClassSpecial Purpose Vehicles/4wd
## VClassSport Utility Vehicle - 2WD ***
## VClassSport Utility Vehicle - 4WD ***
## VClassStandard Pickup Trucks ***
## VClassStandard Pickup Trucks 2WD ***
## VClassStandard Pickup Trucks 4WD ***
## VClassStandard Pickup Trucks/2wd *
## VClassStandard Sport Utility Vehicle 2WD ***
## VClassStandard Sport Utility Vehicle 4WD ***
## VClassSubcompact Cars **
## VClassTwo Seaters ***
## VClassVans ***
## VClassVans Passenger
## VClassVans, Cargo Type ***
## VClassVans, Passenger Type ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 9.856 on 40047 degrees of freedom
## Multiple R-squared: 0.1152, Adjusted R-squared: 0.1145
## F-statistic: 158 on 33 and 40047 DF, p-value: < 2.2e-16sCharger, tCharger
- Definition
- sCharger - if S, this vehicle is supercharged
Remarks
- 6302 tCharger Vehicles
- 796 sCharger Vechiles
- large amount of missing values make these variables useless
## S
## 39285 796## [1] 796## [1] Not Available Hybrid FFV Plug-in Hybrid
## 9 Levels: Bifuel (CNG) Bifuel (LPG) CNG Diesel EV FFV Hybrid ... Plug-in Hybrid## Mode TRUE NA's
## logical 6302 33779##
## Call:
## lm(formula = UCity ~ tCharger, data = vhcls_data)
##
## Coefficients:
## (Intercept) tChargerTRUE
## 24.52 NA## [1] Not Available Diesel Hybrid FFV Plug-in Hybrid
## 9 Levels: Bifuel (CNG) Bifuel (LPG) CNG Diesel EV FFV Hybrid ... Plug-in Hybridyear
Definition
- model year
Remarks
- 1984 to 2019
- Most vehicles have 1984 model
- Year 2010 is a turning point in UCity
## year_cat cylinders_cat
## 1984 : 1964 4 :15475
## 1985 : 1701 6 :13912
## 2018 : 1330 8 : 8645
## 2017 : 1293 5 : 769
## 2015 : 1278 12 : 608
## 2016 : 1257 (Other): 501
## (Other):31258 NA's : 171##
## Call:
## lm(formula = UCity ~ year_cat, data = vhcls_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -21.696 -4.468 -1.176 2.725 197.901
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 22.6191 0.2284 99.053 < 2e-16 ***
## year_cat1985 -0.1511 0.3352 -0.451 0.652228
## year_cat1986 -0.4441 0.3698 -1.201 0.229825
## year_cat1987 -0.9235 0.3664 -2.520 0.011727 *
## year_cat1988 -0.8832 0.3779 -2.337 0.019430 *
## year_cat1989 -1.1094 0.3755 -2.955 0.003130 **
## year_cat1990 -1.2884 0.3836 -3.359 0.000784 ***
## year_cat1991 -1.5512 0.3776 -4.107 4.01e-05 ***
## year_cat1992 -1.6025 0.3788 -4.230 2.34e-05 ***
## year_cat1993 -1.3514 0.3819 -3.539 0.000403 ***
## year_cat1994 -1.4689 0.3955 -3.714 0.000204 ***
## year_cat1995 -1.9031 0.3976 -4.787 1.70e-06 ***
## year_cat1996 -0.9782 0.4297 -2.276 0.022824 *
## year_cat1997 -1.2026 0.4319 -2.784 0.005367 **
## year_cat1998 -1.1134 0.4222 -2.637 0.008366 **
## year_cat1999 -0.8729 0.4152 -2.103 0.035506 *
## year_cat2000 -0.9748 0.4172 -2.336 0.019479 *
## year_cat2001 -0.9791 0.4057 -2.414 0.015804 *
## year_cat2002 -1.5095 0.3965 -3.807 0.000141 ***
## year_cat2003 -1.6279 0.3876 -4.200 2.68e-05 ***
## year_cat2004 -1.7846 0.3787 -4.712 2.46e-06 ***
## year_cat2005 -1.5570 0.3741 -4.162 3.17e-05 ***
## year_cat2006 -1.7979 0.3807 -4.723 2.33e-06 ***
## year_cat2007 -1.8395 0.3783 -4.863 1.16e-06 ***
## year_cat2008 -1.4256 0.3721 -3.832 0.000127 ***
## year_cat2009 -0.8700 0.3723 -2.337 0.019465 *
## year_cat2010 0.1277 0.3801 0.336 0.736853
## year_cat2011 0.9275 0.3779 2.455 0.014105 *
## year_cat2012 1.9746 0.3756 5.258 1.47e-07 ***
## year_cat2013 3.8167 0.3723 10.250 < 2e-16 ***
## year_cat2014 4.3227 0.3688 11.721 < 2e-16 ***
## year_cat2015 4.8909 0.3637 13.447 < 2e-16 ***
## year_cat2016 6.5099 0.3655 17.809 < 2e-16 ***
## year_cat2017 6.8237 0.3624 18.828 < 2e-16 ***
## year_cat2018 6.4955 0.3594 18.075 < 2e-16 ***
## year_cat2019 4.2796 0.5106 8.381 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 10.12 on 40045 degrees of freedom
## Multiple R-squared: 0.06717, Adjusted R-squared: 0.06636
## F-statistic: 82.39 on 35 and 40045 DF, p-value: < 2.2e-16Others
engId
- Definition
- EPA model type index
- Just an index variable. no use for prediction
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0 0 186 8377 4301 69102eng_desc
eng_dscr - engine descriptor;
## Length Class Mode
## 0 NULL NULLid
- Definition: Engine record id
model
- model - model name
Summary Insights
- The dataset has two contrast cases of vehicles. (39829 non EV) and EV
- The median annual petroleum consumption (barrels08) is start to decrease post 2010.
- A strong rise in FFV (dual fuel vehicles) during 2000-2010 period and the a sharp drop to the lowest.
- Premium and Regular fuelType vehicle are top co2 emitters
- cylinders are negatively correlated with UCity. (-0.8157), which indicates the bigger the vehicle the less the miles per gallon. Make sense.
- Vehicles with 4, 6 and 8 cylinders vehicles are getting slightly better on UCity in recent years
- over the years, displ distributions is not changing noticeably.
- youSaveSpend also shows a trend reversal at year 2010 - EV, Hybrid, Plug-in Hybrid and CNG vechicles return positive
- Most common transmission type is Automatic 4 and 4 Speed
- EV, Hybrid and Plug-in Hybrid are least expensive i.t.o fuelCost08
- Over the years, Automatic transmission cars are beoming more and more common
- On broader categorization (Automatic, Manual), Manual transmission does improve UCity 1.43922 as much as.
- Compact, Subcompact and Midsize cars are the top 3 in the data
- Special Purpose Verticia, Vans Passenger are the least
- Most vehicles in the dataset are 1984 model
- Year 2010 is a turning point in UCity