Statistical parameters : a (very) quick guide
Use this when you want to quickly understand the performance of your statistical models or while reviewing a paper.
Coefficient of determination:
Measures the degree of variance in the dependent variables that are measured by an independent variable.
Range of values: 0 to 1
How to interpret: The higher the coefficient of determination, the better the model’s predictive performance.
More information: Link
Pearson’s R
Measures the strength and direction of the relationship between two variables.
Range of values: 0 to 1
How to interpret: The higher the value, the better the model performance.
More information: Link
Root Mean Squared Error (RMSE)
It measures the average difference between values predicted by a model and the actual values. It provides an estimation of how well the model is able to predict the target value (accuracy).
Range of values: N/A
How to interpret: The lower the Root Mean Squared Error value, the better the model is. A perfect model (a hypothetic model that would always predict the exact expected value) would have a Root Mean Squared Error value of 0.
More information: Link
Standard Deviation
A standard deviation (or σ) is a measure of how dispersed the data is in relation to the mean.
Range: N/A
How to interpret: Low standard deviation means data are clustered around the mean, and high standard deviation indicates data are more spread out.
More information: Link
Coefficient of variation (COV)
A ratio between the standard deviation and the mean.
Range of values: N/A
How to interpret: The standard deviation equals the mean if the value equals one or 100%. Values less than one indicate that the standard deviation is smaller than the mean (typical), while values greater than one occur when S.D. is greater than the mean.
The higher the coefficient of variation, the greater the level of dispersion around the mean. It is generally expressed as a percentage.
More information: Link
Mean Squared Error (MSE)
The Mean Squared Error measures how close a regression line is to a set of data points. A larger MSE indicates that the data points are dispersed widely around its central moment (mean), whereas a smaller MSE suggests the opposite.
Range: N/A
How to interpret: A smaller MSE is preferred because it indicates that your data points are dispersed closely around its mean.
More information: Link
Mean Absolute Error (MAE)
Absolute Error is the amount of error in your measurements. It is the difference between the measured value and “true” value.
Range: N/A
How to interpret: The smaller the MAE, the better the model performance.
More information: Link
I’d be more than happy to add more performance metrics, let me know what you want in the comments!