Fits curves with an AR1 or a power-distance correlation model (R.W. Payne).

### Options

`PRINT` = string tokens |
What to print (`model` , `deviance` , `summary` , `estimates` , `correlations` , `fittedvalues` , `accumulated` , `monitoring` , `cparameter` , `cmonitoring` , `cplot` ); default `mode` , `summ` , `esti` , `cpar` |
---|---|

`CURVE` = string token |
Which standard curve to fit (`exponential` , `dexponential` , `cexponential` , `lexponential` , `logistic` , `glogistic` , `gompertz` , `ldl` , `qdl` , `qdq` , `fourier` , `dfourier` , `gaussian` , `dgaussian` ); default `expo` |

`SENSE` = string token |
Sense of a standard curve (`right` , `left` ); default `righ` |

`ORIGIN` = scalars |
Constrained origin for a standard curve; default `*` i.e. not constrained |

`NONLINEAR` = string token |
How to treat nonlinear parameters between groups in standard curves (`common` , `separate` ); default `comm` |

`CALCULATION` = expression structures |
Define a nonlinear model involving explanatory variates and nonlinear parameters; default `*` implies that a standard curve is fitted |

`CONSTANT` = string token |
How to treat the constant (`estimate` , `omit` ); default `esti` |

`FACTORIAL` = scalars |
Limit for expansion of model terms; default 3 |

`POOL` = string token |
Whether to pool ss in accumulated summary between all terms fitted in a linear model (`yes` , `no` ); default `no` |

`DENOMINATOR` = string token |
Whether to base ratios in accumulated summary on rms from model with smallest residual ss or smallest residual ms (`ss` , `ms` ); default `ss` |

`NOMESSAGE` = string tokens |
Which warning messages to suppress (`dispersion` , `leverage` , `residual` , `aliasing` , `marginality` , `vertical` , `df` , `inflation` ); default `*` |

`FPROBABILITY` = string token |
Printing of probabilities for variance and deviance ratios (`yes` , `no` ); default `no` |

`SELECTION` = string tokens |
Statistics to be displayed in the summary of analysis produced by `PRINT=summary` (`%variance` , `%ss` , `adjustedr2` , `r2` , `seobservations` , `dispersion` , `%cv` , `%meandeviance` , `%deviance` , `aic` , `bic` , `sic` ); default `%var` , `seob` |

`SELINEAR` = string token |
Whether to calculate s.e.s for linear parameters when nonlinear parameters are also estimated (`yes` , `no` ); default `no` |

`WEIGHTS` = variate |
Prior weights for the units |

`CPARAMETER` = scalars |
Correlation parameter |

`CPOSITIONS` = variate |
Correlation positions |

`CGROUPS` = factor |
Groupings of correlation positions |

`MAXCYCLE` = scalars |
Maximum number of iterations; default 100 |

`TOLERANCE` = scalars |
Convergence criterion; default 10^{-5} |

### Parameter

`TERMS` = formula |
Terms to be fitted |
---|

### Description

`NLAR1`

allows you to fit curves and nonlinear models to data, such as repeated measurements, where the residuals may follow an AR1 or a power-distance correlation model. The `CPOSITIONS`

option specifies the coordinates of the observations in the direction (e.g. time) along which the correlation model operates. You can also use the `CGROUPS`

option to specify a factor to define groups of observations for the model – the correlation model is then defined only over the observations that belong to the same groups. If you are fitting a standard curve, `CPOSITIONS`

will take the x-variate for the curve as its default, and the group factor (if specified e.g. to define parallel curves) as the default for `CGROUPS`

. `NLAR1`

also allows the data units to have unequal weights, which can be supplied in a variate using the `WEIGHTS`

option.

The parameter phi of the AR1 or power-distance model is estimated within `NLAR1`

, and is assumed to be the same for every group. (Note that the model will be AR1 if the observations are each one unit apart within each group – the power-distance model is the natural extension of the AR1 model to unequally-spaced data; see Method.) You can save the estimated value of phi, in a scalar, using the `CPARAMETER`

option.

Otherwise, `NLAR1`

is used much like `FITCURVE`

or `FIT`

(which are used inside `NLAR1`

to fit the model). `NLAR1`

must be preceded by a `MODEL`

statement. You must also give an `RCYCLE`

statement first if you want to fit a user-defined nonlinear model (using `FIT`

), rather than a standard curve (using `FITCURVE`

). The `MODEL`

statement must have the `WEIGHT`

option set to a symmetrix matrix, which need not have any values defined. `NLAR1`

will set the values according to the distances (`CPOSITIONS`

), groups (`CGROUPS`

) and estimated parameter phi. These values remain set after `NLAR1`

. So you can display or save further output using `RCHECK`

, `RDISPLAY`

, `RGRAPH`

or `RKEEP`

, in the usual way. You could also, for example, use `NLAR1`

to fit a full set of regression terms, and then use `DROP`

to investigate smaller models while still using the phi estimate from the full model. `NLAR1`

has a `TERMS`

parameter to specify the terms to be fitted, like the parameter of `FIT`

and `FITCURVE`

. It also has options `CURVE`

, `SENSE`

, `ORIGIN`

, `NONLINEAR`

, `CALCULATION`

, `CONSTANT`

, `FACTORIAL`

, `POOL`

, `DENOMINATOR`

, `NOMESSAGE`

, `FPROBABILITY`

, `SELECTION`

and `SELINEAR`

which operate like those of `FITCURVE`

and `FIT`

. If the `CALCULATION`

option is unset, then options `CURVE`

, `SENSE`

, `ORIGIN`

, `NONLINEAR`

define which standard curve to fit (using `FITCURVE`

). Alternatively, if `CALCULATION`

is set, those options are ignored, and the expressions specified by `CALCULATION`

define a nonlinear model to be fitted (by `FIT`

).

The `PRINT`

option is also similar, except that it has three additional settings:

`cparameter` |
prints the estimated value of the correlation phi, together with a test for phi=0, |
---|---|

`cmonitoring` |
provides monitoring information for the estimation of phi, |

`cplot` |
plots the likelihood for phi. |

Note, the likelihood values omit some constant terms that depend only on the regression terms. The default is `PRINT=model,summary,estimates,cparameter`

.

The other options control the estimation. The `MAXCYCLE`

option defines the maximum number of iterations (default 100) used to estimate phi, and the `TOLERANCE`

option specifies the convergence criterion i.e. the accurary to which phi is to be estimated (default 10^{-5}).

Options: `PRINT`

, `CURVE`

, `SENSE`

, `ORIGIN`

, `NONLINEAR`

, `CALCULATION`

, `CONSTANT`

, `FACTORIAL`

, `POOL`

, `DENOMINATOR`

, `NOMESSAGE`

, `FPROBABILITY`

, `SELECTION`

, `SELINEAR`

, `WEIGHTS`

, `CMETHOD`

, `CPARAMETER`

, `CPOSITIONS`

, `CGROUPS`

, `MAXCYCLE`

, `TOLERANCE`

.

Parameter: `TERMS`

.

### Method

To estimate phi `NLAR1`

uses procedure `MIN1DIMENSION`

, which calls a procedure `_MIN1DFUNCTION`

, which is loaded automatically with `NLAR1`

. `_MIN1DFUNCTION`

uses the `FITCURVE`

or `FIT`

directives to fit the regression model for a particular value of phi, and then evaluates the likelihood. If standard curves are fitted using `FITCURVE`

for groups of observations, these groups must be independent. Otherwise `FITCURVE`

will give a fault diagnostic. (Thus the default setting for the `CGROUPS`

option with standard curves is the group factor, if one has been specified in the `TERMS`

formula.)

### Action with `RESTRICT`

Restrictions are not allowed.

### See also

Directives: `FITCURVE`

, `FITNONLINEAR`

, `VSTRUCTURE`

.

Procedure: `RAR1`

.

Commands for: Repeated measurements, Regression analysis.

### Example

CAPTION 'NLAR1 example'; STYLE=meta VARIATE [VALUES=5...30] x & [VALUES=1.30,3.55,5.13,6.48,7.85,8.96,9.84,10.91,11.29,11.76,\ 12.12,12.55,12.70,13.14,13.47,13.78,14.01,14.11,14.55,14.71,\ 14.57,14.30,14.67,14.68,15.03,15.00] y SYMMETRIC [ROWS=26] wt MODEL [WEIGHTS=wt] y NLAR1 [CURVE=exponential] x