#include <genesis/utils/math/regression/glm.hpp>
Definition at line 100 of file glm.hpp.
|
| std::vector< double > | betaQ |
| | Vector of parameter estimates (in terms of basis matrix, Xb) (size M). More...
|
| |
| bool | converged = false |
| |
| double | deviance = 0.0 |
| | Deviance. More...
|
| |
| size_t | df_resid = 0 |
| | Residual degrees of freedom. More...
|
| |
| std::vector< double > | fitted |
| | Fitted values (size N). More...
|
| |
| double | null_deviance = 0.0 |
| | Null deviance. More...
|
| |
| size_t | num_iterations = 0 |
| |
| size_t | rank = 0 |
| | Rank of X after regression on strata. More...
|
| |
| std::vector< double > | resid |
| | Working residuals (on linear predictor scale) (size N). More...
|
| |
| double | scale = 1.0 |
| | Scale factor (scalar). More...
|
| |
| std::vector< double > | tri |
| | Upper unit triangular transformation matrix, with Xb - tr.Xb placed in the diagonal (size (M * (M+1)) / 2). More...
|
| |
| std::vector< double > | weights |
| | Weights (size N) More...
|
| |
| std::vector< double > | which |
| | Which columns in the X matrix were estimated (first = 0) (size M). More...
|
| |
| Matrix< double > | Xb |
| | Orthogonal basis for X space (N * M matrix, with N * rank being used). More...
|
| |
◆ betaQ
| std::vector<double> betaQ |
Vector of parameter estimates (in terms of basis matrix, Xb) (size M).
Use glm_estimate_betas() to transform this back into the basis of the original predictors.
Definition at line 150 of file glm.hpp.
◆ converged
◆ deviance
Deviance.
Difference in goodness of fit between the fitted model and the perfect model, quantifying how well the model with predictors explains the variability in the response variable compared to the null model.
Definition at line 174 of file glm.hpp.
◆ df_resid
Residual degrees of freedom.
Definition at line 113 of file glm.hpp.
◆ fitted
| std::vector<double> fitted |
Fitted values (size N).
Definition at line 128 of file glm.hpp.
◆ null_deviance
| double null_deviance = 0.0 |
Null deviance.
Goodness of fit of a model with only the intercept (no predictors) compared to a perfect model, indicating how well the model with no predictors explains the variability in the response variable.
Definition at line 165 of file glm.hpp.
◆ num_iterations
| size_t num_iterations = 0 |
◆ rank
Rank of X after regression on strata.
Definition at line 108 of file glm.hpp.
◆ resid
| std::vector<double> resid |
Working residuals (on linear predictor scale) (size N).
Definition at line 133 of file glm.hpp.
◆ scale
Scale factor (scalar).
Definition at line 118 of file glm.hpp.
◆ tri
Upper unit triangular transformation matrix, with Xb - tr.Xb placed in the diagonal (size (M * (M+1)) / 2).
Definition at line 156 of file glm.hpp.
◆ weights
| std::vector<double> weights |
Weights (size N)
Definition at line 138 of file glm.hpp.
◆ which
| std::vector<double> which |
Which columns in the X matrix were estimated (first = 0) (size M).
Definition at line 143 of file glm.hpp.
◆ Xb
Orthogonal basis for X space (N * M matrix, with N * rank being used).
Definition at line 123 of file glm.hpp.
The documentation for this struct was generated from the following file:
1.8.17