Questions tagged [mse]
MSE stands for Mean Squared Error. It is a measure of the performance of an estimate or prediction, equal to the mean squared difference between the observed values and the estimated / predicted values.
440 questions
0 votes
0 answers
16 views
Plotting Training VS Testing Curve
I am using gradient boosting regressor from scikit-learn with squared error as the loss function. Then i want to plot the training set vs test set curve. Based on what i read, it is used to see the ...
- 11
5 votes
2 answers
341 views
Why can a model with higher MSE still have a higher R² than another model
Two mdoels trained on different training datasets (related but not exactly the same) and tested on the exact same datasets, the model that has higher MSE, still has a higher R², is that possible?
- 131
1 vote
0 answers
76 views
Pointwise bandwidth selection for kernel density estimation
Consider an estimation for $f(x)$ for a function $f$ at a fixed point $x$. We may use kernel density estimator $\hat{f}_n$ to estimate the $f$, then calculate $\hat{f}(x)$. If I understand correctly, ...
- 117
3 votes
1 answer
84 views
Lower bound for MSE, based on sample mean and variance
Short question: For two unknown samples $A$ and $B$ of size $n$, if only their sample mean and sample variances are known, what can be said about $MSE(A,B)$ ? Long version: To be more precise, I ...
- 41
5 votes
3 answers
193 views
Comparison of candidate models obtained with possibly "non-comparable AIC": use RSS, MSE, or adjusted MSE as an alternative instead... or?
My question relates to the comparison of candidate models for which their parameter estimates have been produced with different methods and R packages. As a fictive example, the MASS (continuous ...
- 263
1 vote
0 answers
20 views
How to further reduce MSE/RMSE in CatBoost after tuning, data augmentation, and outlier removal? [duplicate]
I am using the CatBoost model with 100 data points and I have done data augmentation, hyperparameter tuning,cross validation, added isolation forest, and randomizedsearchCV and at the end I could have ...
- 11
0 votes
0 answers
17 views
Why does my GNN-LSTM model fail to generalize with full training data for a spatiotemporal prediction task? [duplicate]
I'm working on a spatiotemporal prediction problem where I want to forecast a scalar value per spatial node over time. My data spans multiple spatial grid locations with daily observations. Data Setup ...
- 43
2 votes
1 answer
117 views
Minimizing MSE for the estimator of population variance
The screenshot below is from the Wikipedia page on Mean Squared Error. Can someone please either help me understand the last highlighted sentence (given the claim is correct), or confirm my suspicion ...
- 137
2 votes
1 answer
171 views
Relationship related to sum of squared deviations
I'm trying to prove this equality, in the context of linear regression: $$ \frac{\sum x_i^2}{n S_{xx}} = \frac{1}{n} + \frac{\bar{x}^2}{S_{xx}} $$ where $$ S_{xx} = \sum (x_i - \bar{x})^2 $$ My ...
- 143
7 votes
1 answer
722 views
Exploiting MSE of binary data for fast search
I have a huge database of binary vectors. Given an incoming vector, I want to find the closest vector in the database in terms of MSE and return the MSE score. So far I have been doing this search ...
- 71
0 votes
0 answers
11 views
What is the best way to set weights for weighted MSE in multi output regression? [duplicate]
I am working on a regression task where the goal is to predict 6 scalar output values from a given input. The input consists of decaying signal data, and the outputs are the parameters of the signal ...
- 11
1 vote
1 answer
87 views
Regarding the definition of unbiasedness and connections to MSE
My impression is that the definition of an unbiased estimator is always made for a single parameter. Is there a reason that the definition cannot be extended for a vector of parameters? For example, ...
- 147
1 vote
0 answers
48 views
Mean squared error of an estimator of samples that are uniformly distributed in (-a,a)
Let $MSE(\hat{\theta}) := \mathbb{E}((\hat{\theta}-\theta)^2)$ be the mean squared error of a statistic $\hat{\theta}$. My question is at the end of the post. The rest is my workout. Let $X_{1},...,...
- 111
1 vote
1 answer
335 views
Way too large MSE for random forest model
I'm currently working on building Random Forest Models in python. My topic is to investigate the Imoportance of specific variables for the accuracy of machine learning to explain the market ...
- 11
4 votes
2 answers
371 views
Deriving MSE($\hat{\beta}$) under Linear regression
I was able to derive the MSE, but there's a part of the derivation which I don't really get. Here's what I got: Facts: $\mathbb{E}(\hat{\beta})=\hat{\beta}\space$ (unbiased estimator) $\text{Cov}(\...
- 391