NAG Library Function Document

nag_all_regsn (g02eac)

void	nag_all_regsn (Nag_OrderType order, Nag_IncludeMean mean, Integer n, Integer m, const double x[], Integer pdx, const char var_names[], const Integer sx[], const double y[], const double wt[], Integer nmod, const char model[], double rss[], Integer nterms[], Integer mrank[], NagError fail)

3 Description

For a set of

k

possible independent variables there are

2^{k}

linear regression models with from zero to

k

independent variables in each model. For example if

k = 3

and the variables are

A

B

and

C

then the possible models are:

(i)	null model
(ii)	$A$
(iii)	$B$
(iv)	$C$
(v)	$A$ and $B$
(vi)	$A$ and $C$
(vii)	$B$ and $C$
(viii)	$A$ , $B$ and $C$ .

nag_all_regsn (g02eac) calculates the residual sums of squares from each of the

2^{k}

possible models. The method used involves a

Q R

decomposition of the matrix of possible independent variables. Independent variables are then moved into and out of the model by a series of Givens rotations and the residual sums of squares computed for each model; see Clark (1981) and Smith and Bremner (1989).

The computed residual sums of squares are then ordered first by increasing number of terms in the model, then by decreasing size of residual sums of squares. So the first model will always have the largest residual sum of squares and the

2^{k}

th will always have the smallest. This aids you in selecting the best possible model from the given set of independent variables.

nag_all_regsn (g02eac) allows you to specify some independent variables that must be in the model, the forced variables. The other independent variables from which the possible models are to be formed are the free variables.

4 References

Clark M R B (1981) A Givens algorithm for moving from one linear model to another without going back to the data Appl. Statist. 30 198–203

Smith D M and Bremner J M (1989) All possible subset regressions using the

Q R

decomposition Comput. Statist. Data Anal. 7 217–236

Weisberg S (1985) Applied Linear Regression Wiley

5 Arguments

1: order – Nag_OrderTypeInput

On entry: the order argument specifies the two-dimensional storage scheme being used, i.e., row-major ordering or column-major ordering. C language defined storage is specified by

order = Nag_RowMajor

. See Section 3.2.1.3 in the Essential Introduction for a more detailed explanation of the use of this argument.

Constraint:

order = Nag_RowMajor

Nag_ColMajor

2: mean – Nag_IncludeMeanInput

On entry: indicates if a mean term is to be included.

$mean = Nag_MeanInclude$: A mean term, intercept, will be included in the model.
$mean = Nag_MeanZero$: The model will pass through the origin, zero-point.

Constraint:

mean = Nag_MeanInclude

Nag_MeanZero

3: n – IntegerInput

On entry:

n

, the number of observations.

Constraints:

$n \geq 2$ ;
$n \geq m$ , is the number of independent variables to be considered (forced plus free plus mean if included), as specified by mean and sx.

4: m – IntegerInput

On entry: the number of variables contained in x.

Constraint:

m \geq 2

5: x[ $\dim$ ] – const doubleInput

Note: the dimension, dim, of the array x must be at least

$\max (1, pdx \times m)$ when $order = Nag_ColMajor$ ;
$\max (1, n \times pdx)$ when $order = Nag_RowMajor$ .

Where

X (i, j)

appears in this document, it refers to the array element

$x [(j - 1) \times pdx + i - 1]$ when $order = Nag_ColMajor$ ;
$x [(i - 1) \times pdx + j - 1]$ when $order = Nag_RowMajor$ .

On entry:

X (i, j)

must contain the

i

th observation for the

j

th independent variable, for

i = 1, 2, \dots, n

and

j = 1, 2, \dots, m

6: pdx – IntegerInput

On entry: the stride separating row or column elements (depending on the value of order) in the array x.

Constraints:

if $order = Nag_ColMajor$ , $pdx \geq n$ ;
if $order = Nag_RowMajor$ , $pdx \geq m$ .

7: var_names[m] – const char *Input

On entry:

var_names [i - 1]

must contain the name of the independent variable in row

i

of x, for

i = 1, 2, \dots, m

8: sx[m] – const IntegerInput

On entry: indicates which independent variables are to be considered in the model.

$sx [j - 1] \geq 2$: The variable contained in the $j$ th column of X is included in all regression models, i.e., is a forced variable.
$sx [j - 1] = 1$: The variable contained in the $j$ th column of X is included in the set from which the regression models are chosen, i.e., is a free variable.
$sx [j - 1] = 0$: The variable contained in the $j$ th column of X is not included in the models.

Constraints:

$sx [j - 1] \geq 0$ , for $j = 1, 2, \dots, m$ ;
at least one value of $sx = 1$ .

9: y[n] – const doubleInput

On entry:

y [i - 1]

must contain the

i

th observation on the dependent variable,

y_{i}

, for

i = 1, 2, \dots, n

10: wt[ $n$ ] – const doubleInput

On entry: optionally, the weights to be used in the weighted regression.

wt [i - 1] = 0.0

, then the

i

th observation is not included in the model, in which case the effective number of observations is the number of observations with nonzero weights.

If weights are not provided then wt must be set to NULL and the effective number of observations is n.

Constraint: if

wt is not NULL

wt [i - 1] = 0.0

, for

i = 1, 2, \dots, n

11: nmod – Integer *Output

On exit: the total number of models for which residual sums of squares have been calculated.

12: model[ $\dim$ ] – const char *Output

Note: the dimension, dim, of the array model must be at least big enough to hold the names of all the free independent variables which appear in all the models. This will never exceed

2^{k} \times m

, where

k

is the number of free variables in the model.

On exit: the names of the independent variables in each model, represented as pointers to the names provided by you in var_names. The model names are stored as follows:

if the first model has three names, i.e., $nterms [0] = 3$ ; then $model [0]$ , $model [1]$ and $model [2]$ will contain these three names;
if the second model has two names, i.e., $nterms [1] = 2$ ; then $model [3]$ , $model [4]$ will contain these two names.

13: rss[ $\max (2^{k}, m)$ ] – doubleOutput

On exit:

rss [i - 1]

contains the residual sum of squares for the

i

th model, for

i = 1, 2, \dots, nmod

14: nterms[ $\max (2^{k}, m)$ ] – IntegerOutput

On exit:

nterms [i - 1]

contains the number of independent variables in the

i

th model, not including the mean if one is fitted, for

i = 1, 2, \dots, nmod

15: mrank[ $\max (2^{k}, m)$ ] – IntegerOutput

On exit:

mrank [i - 1]

contains the rank of the residual sum of squares for the

i

th model.

16: fail – NagError *Input/Output

The NAG error argument (see Section 3.6 in the Essential Introduction).

6 Error Indicators and Warnings

NE_ALLOC_FAIL: Dynamic memory allocation failed.
NE_BAD_PARAM: On entry, argument $⟨value⟩$ had an illegal value.
NE_FREE_VARS: There are no free x variables.
NE_FULL_RANK: Full model is not of full rank.
NE_INDEP_VARS_OBS: Number of requested $x$ -variables $\geq$ number of observations.
NE_INT: On entry, $m = ⟨value⟩$ .
Constraint: $m \geq 2$ .
On entry, $n = ⟨value⟩$ .
Constraint: $n \geq 2$ .
On entry, $pdx = ⟨value⟩$ .
Constraint: $pdx > 0$ .
NE_INT_2: On entry, $pdx = ⟨value⟩$ and $m = ⟨value⟩$ .
Constraint: $pdx \geq m$ .
On entry, $pdx = ⟨value⟩$ and $n = ⟨value⟩$ .
Constraint: $pdx \geq n$ .
NE_INT_ARRAY_ELEM_CONS: On entry, $sx [⟨value⟩] < 0$ .
NE_INTERNAL_ERROR: An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
NE_REAL_ARRAY_ELEM_CONS: On entry, $wt [⟨value⟩] < 0.0$ .

7 Accuracy

For a discussion of the improved accuracy obtained by using a method based on the

Q R

decomposition see Smith and Bremner (1989).

8 Parallelism and Performance

nag_all_regsn (g02eac) is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library.

nag_all_regsn (g02eac) makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.

Please consult the Users' Note for your implementation for any additional implementation-specific information.

9 Further Comments

nag_cp_stat (g02ecc) may be used to compute

R^{2}

and

C_{p}

-values from the results of nag_all_regsn (g02eac).

If a mean has been included in the model and no variables are forced in then

rss [0]

contains the total sum of squares and in many situations a reasonable estimate of the variance of the errors is given by

rss [nmod - 1] / (n - 1 - nterms [nmod - 1])

10 Example

The data for this example is given in Weisberg (1985). The independent variables and the dependent variable are read, as are the names of the variables. These names are as given in Weisberg (1985). The residual sums of squares computed and printed with the names of the variables in the model.

NAG Library Function Documentnag_all_regsn (g02eac)

+− Contents

1 Purpose

2 Specification

3 Description

4 References

5 Arguments

6 Error Indicators and Warnings

7 Accuracy

8 Parallelism and Performance

9 Further Comments

10 Example

10.1 Program Text

10.2 Program Data

10.3 Program Results

NAG Library Function Document

nag_all_regsn (g02eac)