# Car depreciation forecasting models

Learn how car depreciation forecasting models work: age-based decay functions, depreciation curve fitting, regression inputs, and how to read a resale value forecast.

A car depreciation forecasting model estimates how a vehicle's value falls over time. The model links vehicle age and model year to a depreciation curve, then converts that curve into a resale value forecast. Accurate forecasting depends on fitting the curve to real transaction data for the same make, model and trim. 

- A depreciation curve plots remaining value against vehicle age, starting from the original price.
- Age-based decay functions are simple to apply but ignore mileage, condition and market shifts.
- Regression models use several inputs at once and need enough clean data to be reliable.
- Curve fitting quality matters more than model complexity for a useful resale value forecast.
- Always compare a forecast against observed resale values for the same model year before trusting it.
 

## What a depreciation curve shows 

A depreciation curve plots the residual value of a vehicle against its age. The curve starts near the original price and falls as the vehicle ages. Each point on the curve is an estimate of what the car is worth at that vehicle age. 

The shape of a depreciation curve is not a straight line. Value drops faster in the early years and flattens later, so the curve bends. A forecast curve captures that bend and projects it forward beyond the data you have. 

Reading a depreciation curve takes care. The starting price, the trim level and the model year all shift the curve up or down. Two cars of the same age can sit on very different curves for that reason. 

## Age-based decay functions in practice 

An age-based decay function is the simplest kind of car depreciation forecasting model. It applies a fixed rate of loss for each year of vehicle age. You multiply the previous value by a factor to get the next year's estimate. 

The decay function suits a quick estimate when you only know the age and the original price. Its weakness is that it treats every car the same. Mileage, condition and demand are invisible to a pure age-based model. 

A decay function can use a constant rate or a rate that changes with age. A changing rate fits the real shape of car depreciation better. The choice of rate is the main decision when you build one. 

- Constant rate: one annual factor applied at every age
- Declining rate: larger early losses, smaller later losses
- Split rate: different factors for the first years and the rest
- Age brackets: a separate factor for each group of years
 

## Building a regression model from data 

A regression model estimates residual value from several inputs at once. Vehicle age, model year, mileage and trim usually enter the equation. Regression fits coefficients that weight each input by its effect on price. 

Regression needs a clean dataset of observed sales with those inputs recorded. Missing or inconsistent fields weaken the fit. A regression built on mixed models or trims will blur real differences between them. 

The output of a regression is an equation, not a single number. You feed in the inputs of a listing and receive a resale value forecast for that vehicle. Check the model against held-back sales before you rely on it. 

## Fitting the curve to observed values 

Depreciation curve fitting is the step that ties a model to real resale values. You choose a curve shape, then adjust its parameters until it passes close to the observed points. Good fitting keeps the forecast curve honest. 

Fit each make, model and trim separately where the data allows it. Pooling different vehicles produces a curve that matches none of them. Grouping by model year also helps, since design changes reset the pattern. 

Judge a fit by how far the curve sits from the observed points, not by how smooth it looks. A curve that hugs the data too tightly may chase noise. Keep some data aside to test the fitted curve. 

## From forecast curve to resale value 

A resale value forecast turns a fitted curve into a number for a specific car. You locate the vehicle age on the curve and read the residual value. The forecast is only as good as the curve behind it. 

Residual value forecasts shift with market conditions that no curve predicts on its own. Supply of new cars, fuel costs and model refreshes move prices around the fitted line. Treat the curve as a baseline, not a fixed outcome. 

Compare the forecast with current listings for the same model year and trim. A large gap signals that the curve is out of date. Rebuilding the fit with recent sales brings the forecast back in line. 

## Choosing between simple and complex models 

The choice between an age-based model and a regression model depends on your data and your question. A decay function answers a rough estimate with almost no data. Regression answers a detailed question when good data exists. 

Complexity brings its own risk. A regression with many inputs can overfit and forecast poorly on new vehicles. A simple decay function is easy to explain and hard to misread. Match the model to the decision you need to make. 

Start with the simpler model and add inputs only when the data shows they matter. Compare both models on the same held-back sales. The forecast that stays closer to those sales is the one to keep. 
| Model type | Data needed | Strength | Watch out for | 
| Age-based decay function | Age and original price | Fast and easy to explain | Ignores mileage and condition | 
| Regression | Sales with inputs recorded | Uses several factors at once | Can overfit with too many inputs | 
| Fitted curve | Observed resale values | Follows real depreciation shape | Needs refreshing as prices move |  

## Checking a model before you trust it 

A car depreciation forecasting model earns trust through testing on data it did not see. Hold back a portion of your sales and forecast them with the model. Compare the forecast against what those cars actually sold for. 

Testing reveals where the model fails. Some model years and trims may forecast well while others drift. Note the gaps and decide whether the model is good enough for your purpose. 

Rebuild the fit when new sales data arrives or when the market shifts. A model trained on old prices loses accuracy as conditions change. Keep the test data separate so each check stays fair. 

## Car depreciation forecasting questions 

### What is a car depreciation forecasting model? 

A car depreciation forecasting model is a method that estimates how a vehicle's value will fall over time. It links vehicle age and model year to a depreciation curve, then produces a resale value forecast. The model can be as simple as an age-based decay function or as detailed as a regression. 

### What is depreciation curve modeling? 

Depreciation curve modeling is the process of choosing a curve shape and fitting it to observed resale values. The fit adjusts curve parameters until the line sits close to real sales. The result is a forecast curve that projects residual value for future ages. 

### How does a depreciation curve fitting step work? 

Depreciation curve fitting compares a chosen curve to observed resale values for the same vehicle. The parameters are adjusted until the curve passes near those points. Keeping some sales aside lets you test whether the fitted curve predicts new cases well. 

### How accurate is a resale value forecast from a curve? 

A resale value forecast is only as accurate as the curve and data behind it. Forecasts drift when market conditions or model years change. Comparing the forecast with current listings for the same trim shows whether the curve needs rebuilding. 

### Which is better, an age-based decay function or a regression? 

An age-based decay function is easier to apply and explains itself, while a regression uses several inputs such as mileage and trim. Regression suits situations with clean, detailed sales data. A decay function suits a quick estimate when only age and price are known.
