Dark Mode
1985 Imported Auto Risk and Price Data
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Automobile data focusing on the characteristics, market price, and insurance risk associated with imported vehicles from the 1985 model year. The information provided allows for the analysis of vehicle specifications, their assigned insurance risk rating (known as "symboling"), and their subsequent normalised losses in use compared to other cars. This data is structured to facilitate numeric prediction tasks, particularly treating the vehicle price as the primary predicted attribute.
Columns
The dataset contains 26 attributes, categorised as 15 continuous, 1 integer, and 10 nominal types:
- symboling: An assigned insurance risk factor symbol. Values range from -3 (probably quite safe) to +3 (risky).
- normalized-losses: The relative average loss payment per insured vehicle year, adjusted based on the vehicle’s size classification.
- make: Manufacturer (e.g., bmw, honda, toyota, porsche).
- fuel-type: Diesel or gas.
- aspiration: Standard (std) or turbo.
- num-of-doors: Two or four.
- body-style: Hardtop, wagon, sedan, hatchback, or convertible.
- drive-wheels: 4wd, fwd, or rwd.
- engine-location: Front or rear.
- wheel-base: Continuous measurement (86.6 to 120.9).
- length: Continuous measurement (141.1 to 208.1).
- width: Continuous measurement (60.3 to 72.3).
- height: Continuous measurement (47.8 to 59.8).
- curb-weight: Continuous measurement (1488 to 4066).
- engine-type: DOHC, OHC, ROTOR, etc.
- num-of-cylinders: Written out (e.g., four, six, twelve).
- engine-size: Continuous measurement (61 to 326).
- fuel-system: 1bbl, mpfi, idi, etc.
- bore: Continuous measurement (2.54 to 3.94).
- stroke: Continuous measurement (2.07 to 4.17).
- compression-ratio: Continuous measurement (7 to 23).
- horsepower: Continuous measurement (48 to 288).
- peak-rpm: Continuous measurement (4150 to 6600).
- city-mpg: Continuous measurement (13 to 49).
- highway-mpg: Continuous measurement (16 to 54).
- price: The vehicle price, continuous from 5118 to 45400.
Distribution
The original database contained 205 instances. For research and prediction tasks, a cleaner subset of the data is commonly used, consisting of 159 instances where all nominal attributes and entries with missing values have been discarded. The data file typically used treats price as the designated class attribute. The material includes input from 1985 model specifications, insurance manuals, and collision reports.
Usage
This data is highly suitable for predicting the real-valued attributes of automobiles, specifically the price of a car, using methods like instance-based learning or linear regression. It can also be applied to research how vehicle specifications influence the assigned insurance risk rating, or symboling process, and subsequent average loss payments.
Coverage
The data scope is focused specifically on imported car and truck specifications, insurance risk, and loss payments pertaining to the 1985 model year.
License
CC0: Public Domain
Who Can Use It
- Data Scientists and Machine Learning Engineers: For building and testing predictive models for real-valued attributes, such as vehicle price forecasting.
- Actuarial Scientists: To investigate the factors influencing "symboling," the process of adjusting a car's risk factor based on its characteristics and price.
- Automotive Industry Analysts: To study the correlations between physical attributes (length, engine size, horsepower) and market value or insurance risk.
- Students and Researchers: For educational purposes involving multivariate statistical analysis and data cleaning (handling missing values).
Dataset Name Suggestions
- 1985 Imported Auto Risk and Price Data
- Vehicle Insurance Loss Assessment (1985)
- Imported Car Price Prediction Dataset
- Auto Symboling and Loss Data
Attributes
Original Data Source:1985 Imported Auto Risk and Price Data
QUALITYLoading...