g02eff calculates a full stepwise selection from variables by using Clarke's sweep algorithm on the correlation matrix of a design and data matrix, . The (weighted) variance-covariance, (weighted) means and sum of weights of must be supplied.
The routine may be called by the names g02eff or nagf_correg_linregm_fit_stepwise.
The general multiple linear regression model is defined by
is a vector of observations on the dependent variable,
is an intercept coefficient,
is an matrix of explanatory variables,
is a vector of unknown coefficients, and
is a vector of length of unknown, Normally distributed, random errors.
g02eff employs a full stepwise regression to select a subset of explanatory variables from the available variables (the intercept is included in the model) and computes regression coefficients and their standard errors, and various other statistical quantities, by minimizing the sum of squares of residuals. The method applies repeatedly a forward selection step followed by a backward elimination step and halts when neither step updates the current model.
The criterion used to update a current model is the variance ratio of residual sum of squares. Let and be the residual sum of squares of the current model and this model after undergoing a single update, with degrees of freedom and , respectively. Then the condition:
must be satisfied if a variable will be considered for entry to the current model, and the condition:
must be satisfied if a variable will be considered for removal from the current model, where and are user-supplied values and .
In the entry step the entry statistic is computed for each variable not in the current model. If no variable is associated with a test value that exceeds then this step is terminated; otherwise the variable associated with the largest value for the entry statistic is entered into the model.
In the removal step the removal statistic is computed for each variable in the current model. If no variable is associated with a test value less than then this step is terminated; otherwise the variable associated with the smallest value for the removal statistic is removed from the model.
The data values and are not provided as input to the routine. Instead, summary statistics of the design and data matrix are required.
Explanatory variables are entered into and removed from the current model by using sweep operations on the correlation matrix of , given by:
where is the correlation between the explanatory variables and , for and , and (and ) is the correlation between the response variable and the th explanatory variable, for .
A sweep operation on the th row and column () of replaces:
The th explanatory variable is eligible for entry into the current model if it satisfies the collinearity tests: and
for a user-supplied value () of and where the index runs over explanatory variables in the current model. The sweep operation is its own inverse, therefore, pivoting on an explanatory variable in the current model has the effect of removing it from the model.
Once the stepwise model selection procedure is finished, the routine calculates:
(a)the least squares estimate for the th explanatory variable included in the fitted model;
(b)standard error estimates for each coefficient in the final model;
(c)the square root of the mean square of residuals and its degrees of freedom;
(d)the multiple correlation coefficient.
The routine makes use of the symmetry of the sweep operations and correlation matrix which reduces by almost one half the storage and computation required by the sweep algorithm, see Clarke (1981) for details.
Clarke M R B (1981) Algorithm AS 178: the Gauss–Jordan sweep operator with detection of collinearity Appl. Statist.31 166–169
Dempster A P (1969) Elements of Continuous Multivariate Analysis Addison–Wesley
Draper N R and Smith H (1985) Applied Regression Analysis (2nd Edition) Wiley
1: – IntegerInput
On entry: the number of explanatory variables available in the design matrix, .
2: – IntegerInput
On entry: the number of observations used in the calculations.
3: – Real (Kind=nag_wp) arrayInput
On entry: the mean of the design matrix, .
4: – Real (Kind=nag_wp) arrayInput
On entry: the upper-triangular variance-covariance matrix packed by column for the design matrix, . Because the routine computes the correlation matrix from c, the variance-covariance matrix need only be supplied up to a scaling factor.
5: – Real (Kind=nag_wp)Input
On entry: if weights were used to calculate c then sw is the sum of positive weight values; otherwise sw is the number of observations used to calculate c.
6: – Integer arrayInput/Output
On entry: the value of
determines the set of variables used to perform full stepwise model selection, for .
To exclude the variable corresponding to the th column of from the final model.
To consider the variable corresponding to the th column of for selection in the final model.
To force the inclusion of the variable corresponding to the th column of in the final model.
, for .
On exit: the value of indicates the status of the th explanatory variable in the model.
7: – Real (Kind=nag_wp)Input
On entry: the value of the variance ratio which an explanatory variable must exceed to be included in a model.
8: – Real (Kind=nag_wp)Input
On entry: the explanatory variable in a model with the lowest variance ratio value is removed from the model if its value is less than fout. fout is usually set equal to the value of fin; a value less than fin is occasionally preferred.
9: – Real (Kind=nag_wp)Input
On entry: the tolerance, , for detecting collinearities between variables when adding or removing an explanatory variable from a model. Explanatory variables deemed to be collinear are excluded from the final model.
10: – Real (Kind=nag_wp) arrayOutput
On exit: contains the estimate for the intercept term in the fitted model. If , then contains the estimate for the th explanatory variable in the fitted model; otherwise .
11: – Real (Kind=nag_wp) arrayOutput
On exit: contains the standard error for the estimate of , for .
12: – Real (Kind=nag_wp)Output
On exit: the -statistic for the fitted regression model.
13: – Real (Kind=nag_wp)Output
On exit: the mean square of residuals for the fitted regression model.
14: – IntegerOutput
On exit: the number of degrees of freedom for the sum of squares of residuals.
15: – IntegerInput
On entry: if a subroutine is provided by you to monitor the model selection process, set monlev to ; otherwise set monlev to .
16: – Subroutine, supplied by the NAG Library or the user.External Procedure
You may define your own function or specify the NAG defined default function g02efh.
If , monfun is not referenced; otherwise its specification is:
monfun must either be a module subprogram USEd by, or declared as EXTERNAL in, the (sub)program from which g02eff is called. Arguments denoted as Input must not be changed by this procedure.
17: – Integer arrayUser Workspace
18: – Real (Kind=nag_wp) arrayUser Workspace
iuser and ruser are not used by g02eff, but are passed directly to monfun and may be used to pass information to this routine.
19: – IntegerInput/Output
On entry: ifail must be set to , or to set behaviour on detection of an error; these values have no effect when no error is detected.
A value of causes the printing of an error message and program execution will be halted; otherwise program execution continues. A value of means that an error message is printed while a value of means that it is not.
If halting is not appropriate, the value or is recommended. If message printing is undesirable, then the value is recommended. Otherwise, the value is recommended. When the value or is used it is essential to test the value of ifail on exit.
On exit: unless the routine detects an error or a warning has been flagged (see Section 6).
6Error Indicators and Warnings
If on entry or , explanatory error messages are output on the current error message unit (as defined by x04aaf).
Errors or warnings detected by the routine:
On entry, .
On entry, ; .
On entry, .
On entry, .
Constraint: or .
On entry, .
On entry, .
On entry, .
On entry, .
Constraint: , or , for .
On entry, , for all .
Constraint: there must be at least one free variable.
On entry at least one diagonal element of .
Constraint: c must be positive definite.
The design and data matrix is not positive definite, results may be inaccurate. All output is returned as documented.
All variables are collinear, no model to select.
An unexpected error has been triggered by this routine. Please
See Section 7 in the Introduction to the NAG Library FL Interface for further information.
Your licence key may have expired or may not have been installed correctly.
See Section 8 in the Introduction to the NAG Library FL Interface for further information.
Dynamic memory allocation failed.
See Section 9 in the Introduction to the NAG Library FL Interface for further information.
g02eff returns a warning if the design and data matrix is not positive definite.
8Parallelism and Performance
Background information to multithreading can be found in the Multithreading documentation.
g02eff is not threaded in any implementation.
Although the condition for removing or adding a variable to the current model is based on a ratio of variances, these values should not be interpreted as -statistics with the usual interpretation of significance unless the probability levels are adjusted to account for correlations between variables under consideration and the number of possible updates (see, e.g., Draper and Smith (1985)).
g02eff allocates internally of real storage.
This example calculates a full stepwise model selection for the Hald data described in Dempster (1969). Means, the upper-triangular variance-covariance matrix and the sum of weights are calculated by g02buf. The NAG defined default monitor function g02efh is used to print information at each step of the model selection process.