!
← Back to all models

AVM Model Details: Kingston, ON

Detailed performance metrics for our automated valuation models in this location.

Sale price estimation model

This model estimates the sale price of properties in this location based on historical data and market trends.

Key Performance Metrics

Median Absolute Percentage Error (M. MAPE)7.8%Half of our predictions are within this percentage of the actual value
Testing R² Score0.770How much of the price variation our model explains
Root Mean Square Error$150505Average prediction error in dollars
Mean Absolute Error$84648Average absolute difference between predicted and actual

Accuracy Distribution

Percentage of predictions within various error ranges

Within 5%34.1%
Within 10%59.3%
Within 15%73.5%
Within 20%81.8%

Prediction Bias Analysis

Does the model tend to overestimate or underestimate?

Mean Bias-$7271Average prediction bias (+ means overestimate)
Median Bias+$930Typical prediction bias
Over-prediction Rate50.5%Percentage of predictions that are above the actual value
Under-prediction Rate49.5%Percentage of predictions that are below actual value

Error Distribution Percentiles

The absolute error amount at different percentiles - for example, '90th percentile: $50,000' means 90% of predictions have errors less than $50,000

Percentile10th25th50th75th90th95th
Error Amount$9336$21419$44321$95681$194388$304040

Model Configuration & Training

Training Samples3368
Testing Samples842
Training R²0.870
Testing R²0.770
Last Updated2025-07-03

Additional Performance Metrics

Mean Absolute Percentage Error12.8%Average percentage error across all predictions
Forecast Standard Deviation$150330Spread of prediction errors
Coefficient of Variation23.7%Relative variability of predictions
10th Percentile Error %1.5%10% of predictions have error less than this
90th Percentile Error %28.4%90% of predictions have error less than this

List price estimation model

This model estimates the list price of properties in this location based on historical data and market trends.

Key Performance Metrics

Median Absolute Percentage Error (M. MAPE)8.1%Half of our predictions are within this percentage of the actual value
Testing R² Score0.750How much of the price variation our model explains
Root Mean Square Error$169161Average prediction error in dollars
Mean Absolute Error$95188Average absolute difference between predicted and actual

Accuracy Distribution

Percentage of predictions within various error ranges

Within 5%33.8%
Within 10%56.7%
Within 15%71.5%
Within 20%81.0%

Prediction Bias Analysis

Does the model tend to overestimate or underestimate?

Mean Bias-$7980Average prediction bias (+ means overestimate)
Median Bias+$815Typical prediction bias
Over-prediction Rate50.5%Percentage of predictions that are above the actual value
Under-prediction Rate49.5%Percentage of predictions that are below actual value

Error Distribution Percentiles

The absolute error amount at different percentiles - for example, '90th percentile: $50,000' means 90% of predictions have errors less than $50,000

Percentile10th25th50th75th90th95th
Error Amount$8931$20314$50674$105404$221758$341042

Model Configuration & Training

Training Samples3368
Testing Samples842
Training R²0.850
Testing R²0.750
Last Updated2025-07-03

Additional Performance Metrics

Mean Absolute Percentage Error13.4%Average percentage error across all predictions
Forecast Standard Deviation$168972Spread of prediction errors
Coefficient of Variation25.2%Relative variability of predictions
10th Percentile Error %1.6%10% of predictions have error less than this
90th Percentile Error %27.6%90% of predictions have error less than this

Monthly rent estimation model

This model estimates the monthly rent of properties in this location based on historical data and market trends.

Key Performance Metrics

Median Absolute Percentage Error (M. MAPE)8.4%Half of our predictions are within this percentage of the actual value
Testing R² Score0.460How much of the price variation our model explains
Root Mean Square Error$488Average prediction error in dollars
Mean Absolute Error$360Average absolute difference between predicted and actual

Accuracy Distribution

Percentage of predictions within various error ranges

Within 5%24.8%
Within 10%56.9%
Within 15%66.7%
Within 20%81.0%

Prediction Bias Analysis

Does the model tend to overestimate or underestimate?

Mean Bias-$48Average prediction bias (+ means overestimate)
Median Bias-$18Typical prediction bias
Over-prediction Rate47.1%Percentage of predictions that are above the actual value
Under-prediction Rate52.9%Percentage of predictions that are below actual value

Error Distribution Percentiles

The absolute error amount at different percentiles - for example, '90th percentile: $50,000' means 90% of predictions have errors less than $50,000

Percentile10th25th50th75th90th95th
Error Amount$51$135$252$511$887$1021

Model Configuration & Training

Training Samples609
Testing Samples153
Training R²0.900
Testing R²0.460
Last Updated2025-07-03

Additional Performance Metrics

Mean Absolute Percentage Error12.7%Average percentage error across all predictions
Forecast Standard Deviation$486Spread of prediction errors
Coefficient of Variation17.3%Relative variability of predictions
10th Percentile Error %2.1%10% of predictions have error less than this
90th Percentile Error %27.5%90% of predictions have error less than this