Solved
Contributor
Posts: 64

# Mean Squared Error vs Average Squared Error

In E-Miner, I see 2 selections in the Model Comparison node:

- Average Squared Error

- Mean Squared Error

What is the difference?

Searching the HELP does not yield any decent description:

----->8  snip from HELP file ------------------------------
The Selection Statistic choices are as follows:
Default — The default selection uses different statistics based on the type of target variable and whether a profit/loss matrix has been defined.
If a profit/loss matrix is defined for a categorical target, the average profit or average loss is used.
If no profit/loss matrix is defined for a categorical target, the misclassification rate is used.
If the target variable is interval, the average squared error is used.
Akaike's Information Criterion — chooses the model with the smallest Akaike's Information Criterion value.
Average Squared Error — chooses the model with the smallest average squared error value.
Mean Squared Error — chooses the model with the smallest mean squared error value.

----->8  snip from HELP file ------------------------------

Accepted Solutions
Solution
‎07-15-2016 12:21 AM
Contributor
Posts: 64

## Re: Mean Squared Error vs Average Squared Error

I answered my own question. In a different part of the HELP section is this description:

In linear models, statisticians routinely use the mean squared error (MSE) as the main measure of fit. The MSE is the sum of squared errors (SSE) divided by the degrees of freedom for error. (DFE is the number of cases less the number of weights in the model.) This process yields an unbiased estimate of the population noise variance under the usual assumptions.
For neural networks and decision trees, there is no known unbiased estimator. Furthermore, the DFE is often negative for neural networks. There exist approximations for the effective degrees of freedom, but these are often prohibitively expensive and are based on assumptions that might not hold. Hence, the MSE is not nearly as useful for neural networks as it is for linear models. One common solution is to divide the SSE by the number of cases N, not the DFE. This quantity, SSE/N, is referred to as the average squared error (ASE).

All Replies
Solution
‎07-15-2016 12:21 AM
Contributor
Posts: 64

## Re: Mean Squared Error vs Average Squared Error

I answered my own question. In a different part of the HELP section is this description:

In linear models, statisticians routinely use the mean squared error (MSE) as the main measure of fit. The MSE is the sum of squared errors (SSE) divided by the degrees of freedom for error. (DFE is the number of cases less the number of weights in the model.) This process yields an unbiased estimate of the population noise variance under the usual assumptions.
For neural networks and decision trees, there is no known unbiased estimator. Furthermore, the DFE is often negative for neural networks. There exist approximations for the effective degrees of freedom, but these are often prohibitively expensive and are based on assumptions that might not hold. Hence, the MSE is not nearly as useful for neural networks as it is for linear models. One common solution is to divide the SSE by the number of cases N, not the DFE. This quantity, SSE/N, is referred to as the average squared error (ASE).
☑ This topic is solved.