The latest HTML and trim groups are saved in this browser. Clear removes the HTML and results while keeping your trim groups.
Trim colors match the price comparison plot and use your current trim groups. Show or hide trims below; visibility affects only this plot and its trend line.
Model listing price from the features you choose. Results use the currently filtered listings, so table filters are reflected here automatically.
Hybrids are recognized by Hybrid, HEV, or PHEV in the vehicle name. When Hybrid is a separate feature, that wording is removed from trim names. The Hybrid predictor is disabled when hybrids are excluded.
Give terms the same group name to analyze them together. Edit a group directly, or select terms and group them below. Edits are saved as you type and applied to the model when you leave the field. Counts here include all pasted listings.
| Select | Trim term | Vehicles | Group name |
|---|
Trim is inferred from the unique part of each vehicle name after removing its year and the shared beginning of the names in your input. Names with no remaining suffix use “Unspecified”. If names have no shared beginning, the full name without the year is used. The baseline group has no separate trim coefficient; other groups are compared with it. This avoids redundant trim columns with the intercept. Other redundant features can still prevent a fit.
Choose a split year to compare coefficients for that model year and below versus the following year and above. Both fits use the selected features, trim groups, baseline, and regression filters. The full-sample model still supplies the price predictions and plots.
This is an ordinary least squares (OLS) multiple linear regression. It estimates price while holding every other selected feature constant. That “all else equal” qualification is what makes these coefficients different from the single trend line above.
Predictions use the full regression above. By default, rows sort from least to greatest by actual price − predicted price: negative values mean the listing costs less than predicted, and positive values mean it costs more. Click any column heading to sort. Filter text, choose multiple locations, or set numeric minimums and maximums below each heading. Comparison filters affect only this table; use the Listings filters to change the fitted model. Missing values sort last.
Each point is a vehicle with an actual and predicted price in the comparison table above. The dashed line marks equal prices. Hover for vehicle details; click a point to pin its tooltip, then use the listing link.
Colors use the same trim groups as the first plot. Trim visibility here affects only this plot.